" Half- MUON" DIY SOTA-wannabe KEF coax based

[Stoo]

Excuse me if it would be partial repetition from other posts because KEF speakers get discussed here quite extensively but it seems this essence of my question is different and not specifically discussed anywhere.
Basically I am looking at building a DIY system functionally similar to KEF LS60 (or KEFLS50 WII + dual KC62) but with the looks towards half height skinnier KEF MUON.
Why bother not not just buy either of them - other than trying to get same for "half price", result could be potentially better (at least for my specific purpose) than the LS60/50+KC62 .

I happen to already have 2 x SB Acoustics sb34nrxl75-8 https://hificompass.com/en/speakers/measurements/sbacoustics/sb-acoustics-sb34nrxl75-8
They are in oversized closed boxes abt 70ltr each - the (sub) woofer that i built earlier. As can be seen from measurements on hificompass this driver has very low distortion in the low end and whilst has OKish Xmax of+-10mm and surface area over 500cm2 to cover subbass, it is also relatively light moving mass (less than half of a typical subwoofer driver this size) and strong motor of 17Tm which allows good impulse response and fast transient so makes it good up to about 300-400Hz.

Now reference to half-Muon is half joking, but actually has a meaning in terms of being quite fat and rounded cabinet so to have enough volume for SB34 and have relatively directionless DIFFRACTION-less wide baffle to support the coax driver.

Crossover with intended points of W-M 200-300Hz and M-T 1900-2100Hz (LR 2 or LR4 to be defined by trial) will be active DSP based (i have miniDSP 4x10HD but planning to upgrade to a newer version with better spec i.e. miniDSP Flex which has better SINAD and FIR LP/HP filters.
Amp will have to be 6channel solution, i.e. 4channel Audiophonics hypex MP252 plus my good old NAD C272 to drive the (sub) woofers for which it seems to work really well.

Now the actual question I have is about getting the heart of the system - the KEF coax.. Why I want it and not any other midwoofer + tweeter is because I happen to listen to music in standing position and moving around the room so had inevitably had problems of getting into the crossover lobing nulls - thus abruptly losing the right tone of the music and feel that using coaxial driver is the best way to minimize this problem.
Since most of the coaxial available to DIYers like Seas or SB ones seems to be rather inferior to KEF Uniqs, which are unfortunately not sold to DIY/OEM.. so will have to be harvested from a complete speaker.
From consideration of cost and also not to destroy already good speaker for sake of experimental system (which may end up not as good as I hope) I decided to not consider any of Reference or R-series as a potential donors. Which leaves most likely candidates of KEF Q350 or Q150 and... LS50Meta, possibly utilizing the latter together with the cabinet by intergrading it into a bigger extended baffle maintaining similar curvature (and of course ditching the passive crossover)..

So the guess the ACTUIAL questions are:
1) is this a completely crazy idea? )
I have experience of building several DYI speakers (open baffles and regular) with active amplification and miniDSP crossover but never had patience to give them looks acceptable for living room and they all were fussy for placement and listening spot so trying to avoid those limitations this time and finally i have proper workshop to do decent cabinets.

2) would it be possible to get Q150 or Q350 driver to perform close to the LS50Meta one for the intended range from 200-300Hz up?
Obviously the Q series driver are worse primarily due to steel basket and simpler motor (not sure if lack of Meta makes a real difference) but with magnet bracing basket resonance can be largely minimized and without low frequency modulations motor deficiencies may not matter so much..
Would be good to know peoples thoughts on such project... Thanks in advance

 

[hex168]

Don't know where you are located, but you could try to find a distributor for SICA coaxes. (Take a look at the Sigberg Audio threads.)

 

[Stoo]

OK, its rather quarter than half Muon.. and still not quite. But conceptually design layout would be smth like this (ignore the colours of the bigger cabinet):

 

[Stoo]

Don't know where you are located, but you could try to find a distributor for SICA coaxes. (Take a look at the Sigberg Audio threads.)

To me SICA inherent downside if fabric dome.. I had tried quite a few different speakers and always ended up preferring hard dome tweeters. Also various KEF speaekrs are available in stores here (Dubai) so it is possible to get good enough idea on the kind of sound you gonna get building up on them..

 

[behappybevegan]

Maybe this is the driver you're looking for?

Haut parleur KEF R11n R7

Haut-parleurs, boomer, woofer,uniq, uni-q, medium, tweeter KEF, centre technique agrée par KEF depuis plus de 20 ans

setelec-shop.fr

 

[dwkdnvr]

There are some projects over on diyaudio exploring similar ideas - LS60/mini-blade inspired designs in various stages of maturity.

I've been mulling over similar ideas. I have a 'spare' Q150 around (using the other as a center with my R3s), and have thought about using the Tang Band W5 subs to create an LS60ish design. It's not really possible to DIY something quite as small/narrow as the actual LS60 (at least not easily), but should be close. It's a very slowly moving project, though.

The source that @behappybevegan mentions does seem to be valid, although it wouldn't surprise me if it gets shut down if usage becomes widespread in DIY designs - that's what happened last time folks started using the replacement drivers. (although one was semi-commercial which might have been the difference). In my case I'm not sure I'm considering trying to make these a fully realized 'in production' project, so the Q150 driver is fine for exploring. Given that I have R3s though, I have considered whether I might be able to order spare drivers through service if I decided to try to make them a finished project.

 

[fluid]

Since most of the coaxial available to DIYers like Seas or SB ones seems to be rather inferior to KEF Uniqs, which are unfortunately not sold to DIY/OEM.. so will have to be harvested from a complete speaker.

Not completely true, have a look at the first post in this thread
https://www.audiosciencereview.com/forum/index.php?threads/kef-r-series-dual-concentric.29383/

1) is this a completely crazy idea? )

I don't see that there is much to be gained over placing the sub in a separate box next to the bookshelf. Making a baffle that doesn't make things worse trying to integrate those two is going to be hard.

2) would it be possible to get Q150 or Q350 driver to perform close to the LS50Meta one for the intended range from 200-300Hz up?
Obviously the Q series driver are worse primarily due to steel basket and simpler motor (not sure if lack of Meta makes a real difference) but with magnet bracing basket resonance can be largely minimized and without low frequency modulations motor deficiencies may not matter so much..
Would be good to know peoples thoughts on such project... Thanks in advance

The later more expensive drivers from the Reference and Meta do have improvements that are worthwhile if you can get you hands on them. Crossing over actively to a sub in of itself would go a long way to making any of coaxes perform better.

 

[abdo123]

What’s the goal behind this project? There are already commercial solutions cheaper. (KEF KC62)

 

[sigbergaudio]

To me SICA inherent downside if fabric dome.. I had tried quite a few different speakers and always ended up preferring hard dome tweeters. Also various KEF speaekrs are available in stores here (Dubai) so it is possible to get good enough idea on the kind of sound you gonna get building up on them..

The tweeter in that driver is actually extremely competent, good dispersion, linear response and with low distortion, so I doubt you would be disappointed if you went with that. We didn't see significant improvement comparing with much more expensive coaxes.

 

[thewas]

The drivers of most coaxials unit are usually not really their limiting factor but their integration into one unit in terms of directivity smoothness where the current Genelec and KEF ones are unrivalled.

 

[sigbergaudio]

Here's the sica coax in our SBS.1 for reference:

 

[Ilkless]

The drivers of most coaxials unit are usually not really their limiting factor but their integration into one unit in terms of directivity smoothness where the current Genelec and KEF ones are unrivalled.

Yes, and the modulation of the tweeter, and the on-axis null.

 

[Stoo]

Not completely true, have a look at the first post in this thread
https://www.audiosciencereview.com/forum/index.php?threads/kef-r-series-dual-concentric.29383/

I don't see that there is much to be gained over placing the sub in a separate box next to the bookshelf. Making a baffle that doesn't make things worse trying to integrate those two is going to be hard.

The later more expensive drivers from the Reference and Meta do have improvements that are worthwhile if you can get you hands on them. Crossing over actively to a sub in of itself would go a long way to making any of coaxes perform better.

Thanks for pointing out the other thread I missed it in my read-thru this forum, it gives a lot of valuable "food for thought".

Of course adding sub to almost any small/medium speaker (incl. KEF) makes them better.. but it is not what i am trying to do here:
The mentioned SB 34cm driver (two of which I happen to have already) is a very low distortion WOOFER with decent sensitivity >90dB (yet still with good subbass capability). The famous "there is no replacement for displacement" is still true for the mid/upper bass range if you want to have good headroom and not chop off the dynamic peaks especially obvious on i.e. drumkit or BSO which always struggle to sounds realistic on any small speaker. To achieve that one need either good big woofer or multiple smaller woofers (and those need to be rare beasts with good Xmax and low distortion in lower frequencies.. like Purifi units). Probably using 3-4 Purifi per side woudl be a better solution for the same goal but it is a significant costs, esp compared to drivers that I already have (=zero cost to me).

On other hand there is not a single subwoofer that i heard which was capable being crossed above 70-80Hz without mudding up the sound. I briefly heard KC62 and it didn't seem to be much better in this regard, I also found it only sounds impressive for it's size (similarly to Devialet Phantoms) otherwise really nothing special.

 

[Stoo]

The tweeter in that driver is actually extremely competent, good dispersion, linear response and with low distortion, so I doubt you would be disappointed if you went with that. We didn't see significant improvement comparing with much more expensive coaxes.

I don't doubt it's competence but it's still fabric, buit couple speakers with Scan-Speak D2608/913000 which has apparently low distortion and good reviews and wide usage (incl. from i.e. Troels Gravesen).. I tried in baffles of various size/shape even in waveguide, various crossover points/slopes and EQ... never felt satisfied. Same for Seas TDC/TDF and some Vifas. Same for various commercial speakers I had and/or auditioned, i.e. Wharfedale, Elacs, Jamo.. All had the same kind of shortcomings, smearing the naturally "metallic" sounds making them vague and unreasilstic.
And the moment I switched to hard domes (Focal Chorus) and especially beryllium (Scanspeaks tweeters that i had for a while)- things started coming right, albeit in case of Focal at the cost of increased sibilance.

 

[Stoo]

Maybe this is the driver you're looking for?

Haut parleur KEF R11n R7

Haut-parleurs, boomer, woofer,uniq, uni-q, medium, tweeter KEF, centre technique agrée par KEF depuis plus de 20 ans

setelec-shop.fr

Thanks a lot, that's a "goldmine"! Really makes my project a lot easier(more affordable) but also gives it a different angle

 

[fluid]

Of course adding sub to almost any small/medium speaker (incl. KEF) makes them better.. but it is not what i am trying to do here:
The mentioned SB 34cm driver (two of which I happen to have already) is a very low distortion WOOFER with decent sensitivity >90dB (yet still with good subbass capability). The famous "there is no replacement for displacement" is still true for the mid/upper bass range if you want to have good headroom and not chop off the dynamic peaks especially obvious on i.e. drumkit or BSO which always struggle to sounds realistic on any small speaker. To achieve that one need either good big woofer or multiple smaller woofers

I think you perhaps misunderstood what I was trying to suggest, as long as the woofer is within 1/4 wavelength of the frequency you want to use it to there is no detriment to having the woofer in a separate box. On the other hand trying to make a good baffle and enclosure for two very differently sized drivers is not at all easy. As an example for 300Hz as long as you are within ~28cm CTC the drivers will behave as one apart from diffraction effects. Lower that to 200Hz and the distance moves to 43cm.

If you really want them in the same enclosure then I would look at mounting the woofer to the side, but that has it's own pro's and con's.

You can see some BEM sims of side mounted woofers I made here
https://www.diyaudio.com/community/threads/tachi-coaxial-point-source-bem-modelling.386711/

 

[Stoo]

I think you perhaps misunderstood what I was trying to suggest, as long as the woofer is within 1/4 wavelength of the frequency you want to use it to there is no detriment to having the woofer in a separate box. On the other hand trying to make a good baffle and enclosure for two very differently sized drivers is not at all easy. As an example for 300Hz as long as you are within ~28cm CTC the drivers will behave as one apart from diffraction effects. Lower that to 200Hz and the distance moves to 43cm.

If you really want them in the same enclosure then I would look at mounting the woofer to the side, but that has it's own pro's and con's.

You can see some BEM sims of side mounted woofers I made here
https://www.diyaudio.com/community/threads/tachi-coaxial-point-source-bem-modelling.386711/

Placing woofer and and mid/hi into same enclosure is not the goal in itself. Maybe my sketch was misleading as it was based on LS50 put on the side and integrated into a bigger cabinet..
But since there is possibility to buy KEF raw drivers from https://setelec-shop.fr/ this gives a lot more playroom on cabinet layout and construction.

My constraint is whilst the room is decent size 50m2 but placement options are quite limited - pretty much has to be against the wall, in which case i believe not too deep but wide cabinet has an advantage compared to narrow but deep one (i.e. less comb filtering in upper bass/midrange due to wall bounce interference). Pulling narrow speaker further like 0.5-1m into the room may result in better spaciousness and soundstage but not an option for me.

 

[fluid]

My constraint is whilst the room is decent size 50m2 but placement options are quite limited - pretty much has to be against the wall, in which case i believe not too deep but wide cabinet has an advantage compared to narrow but deep one (i.e. less comb filtering in upper bass/midrange due to wall bounce interference). Pulling narrow speaker further like 0.5-1m into the room may result in better spaciousness and soundstage but not an option for me.

That does limit you quite severely, my own speakers are 0.5m (cabinet front) from the front wall by design. Getting closer than that without causing problems is tricky and drivers with limited depth are needed, the KEF coaxes are quite deep, never mind the 12" woofer. Wider and thinner is better for close or near wall placement but without being careful there can be a pretty big SBIR null created.

 

[Stoo]

0.5m (cabinet front) from the front wall

Whilst such distance still works OK for bass, but for upper bass/lower midrange that could be problematic placement in case of narrow baffle speakers due to early reflections and comb filtering.
Nonetheless it is probably possible to make it work to one's satisfaction (because this is how most people seems to actually have their speakers) but having tried in my case - what works much better is placement either "freestanding" (meaning at least 1m into the room, ideally with separate bass unit against the wall) or wide baffle close to the wall (which effectively approximates flush mounting into the wall). Now if we talk abt depth of the cabinet itself, realistic minimum is 25-30cm for bass section with bigger unit and possibly 15-20cm for midrange. Making it less is difficult because by the time we deduct baffle and wall thicknesses, felt etc we have little room left for the driver to breath.

 

[fluid]

Whilst such distance still works OK for bass, but for upper bass/lower midrange that could be problematic placement in case of narrow baffle speakers due to early reflections and comb filtering.

What works and doesn't depends a lot on the speaker design itself and it's directivity. Generalizations are just that and are inevitable when the designer doesn't know the placement. You have an advantage there as you know where you will put them and can tailor your design to suit.