3-way active standmounts with coax upper end & Other UniQ adventures

I have been thinking about a project a project like this for some time, so why not start a thread to get it going. The plan is to build fairly large standmount speakers using a coaxial mid-high and a 10”woofer, so quite much inspired by Genelec 8260.

pd8260a01-1024x768.jpg


Objective 1: Learning and having fun. Learn new design, simulation, and fabrication skills. Getting more experience with Ath, VituixCAD, and Fusion360. And on the fabrication side the goal is to use all the cool toys available at the Uni FabLab (3D scanning, 3D printing, laser cutting, CNC…).

Objective 2: Controlled directivity and low diffraction using waveguides and baffle shaping.

Objective 3: The result should sound in our compact living room better than my current speakers, SP38/13 by Troels Gravesen. This rather ambitions, but let’s see…


Electronics: miniDSP 4x10HD and TPA3255 amps with 4x150W and 2x300W.

Initial thoughts on driver selection: Kef Q100 drivers would be quite safe choice for this application, but they are not so easy to obtain these days. The newer Q150 model is easy to come by, but I have understood that it uses inferior drivers (stamped steel basket etc). The other alternative I have in mind is Kartesian Cox120 from TLPH. There’s not much info or experiences available for this driver though. It’s a rather small 4.5” coax but the inverted surround of the cone would allow seamless integration with an additional waveguide structure. I think I will run some simulations with Ath to find out the tweeter polar response with and without an additional waveguide. I have excluded SEAS King Coax because of the price.

There are much more good candidates for the lower end. The woofer should work in ~30 l sealed enclosure and allow crossing in the 250-400Hz in order that relieve the midcone from the bass duties. Options include SEAS L26ROY, Dayton RSS265HF, SB wo24p, SB29NRX75… I suppose this is a compromise between nice upper bass and capacity at the lower end (woofer vs subwoofer). My amps like low impedances so 4 ohm models preferred in that respect.

Edit: A pair Kef Q100 drivers was aquired for this project.

Here is a list of all coaxial drivers linked to linked to this thread:

Kartesian Cox120
Coaxial speaker Kartesian Cox120_vHP, 8 ohm, 120 mm
Kef Q100 driver SP1587 5” Uni-Q, 4 Ohm measurements
Kef Q100 Speaker Drive Unit Testing

Omnes Audio CX 3.1
Die Klangmanufaktur | Omnes Audio CX 3.1 | Lautsprecher Manufaktur
Omnes Audio Exclusive 8AL Monitor
Omnes Audio Exclusiv 3/8

Seas MR18REX/XF

Seas MR18REX/XF (H1699) 7" Midrange Coaxial: Madisound Speaker Components
DIY speaker by Juhazi

Seas L12RE/XFC
https://seas.no/index.php?option=com_content&view=article&id=106:h1602-0406-l12rexfc&catid=52&Itemid=464
Desing and crossover from Seas/Madisound:http://www.madisound.com/pdf/L12RE_XFC_AppNote.pdf
There is a commercial 2-way that uses this driver and got rave reviews in Stereophile:
Trenner & Friedl Sun loudspeaker | Stereophile.com

Sica 5.5 C 1.5 CP
Boutique Haut-parleurs & Audio DIY
DIY speaker project based on Sica coaxials: Neue-Interessante-Sica-5-5-und-6-5-Zoll-Coaxe-mit-Kalotten-HT

Tangband Q6-2313
W6-2313 - 6.5” Paper Coaxial Full Range - TB SPEAKER CO., LTD.
 
Last edited:
3-way active standmounts with coax upper end

You can check Ebay for Tannoy in-wall/ceiling speakers, just make sure you pick one with the tulip style tweeter.

I love the idea of the L26ROY for the bottom end.


Why Tulip only? The later “Omnimagnet” Dual Concentric is in some respects an improvement. It has more even midrange dispersion. The ring radiator tweeters do lack the extension of the older domes though.

Now ICT…yeah, avoid.
 
Tannoy drivers are difficult to find from stores and eBay from other side of globe is a bit risky.

I have 3-way a project with MR18 and it is not the smoothest coaxial around...

I might try the Kartesian 6,5" or Tangband 6.5" (they are pretty)

Woofer choice depends a bit on box size, LROY fits in small box, XXLS10 or RSS225 might be nice too.

I recommend tower for L/R because more choices with drivers and easy to double the drivers to get lower distortion and lower tuning (or sealed). Subwoofer is not needed for music then, and hardly even for movies. But if you have good sub(s) already... and dsp channels to set nice slopes and handle delay.
 
Kartesian Cox165 doensn't seem very smooth based on the measurement data.
Kartesian Cox165_vPA reviews?

Meanwhile, I did some simulations for Kartesian Cox120 with Ath. First attachment is simulated for what I estimated as the geometry of the tweeter based on the driver drawings and pictures (not precise). The second attechment is the result for an imaginary dome that meets the cox120 cone in a nice angle without any steps or gaps. Both figures show results with and without the additional OS-SE waveguide shown in post #2. I used circular symmetry and infinite baffle for all simulations. The results from rough 3D simulations are similar, but I didn't have time to do detailed testing without circular symmetry. The WG seems to improve tweeter directivity by an octave or so. The response doesn't look very nice for the highest freuqencies. The result is cleaner for the "optimal" dome. I think I would need more accurate data for the tweeter geometry in order to get anything reliable for the highest frequencies.
 

Attachments

  • cox120.jpg
    cox120.jpg
    80.3 KB · Views: 1,237
  • dome.jpg
    dome.jpg
    75.3 KB · Views: 1,112
Last edited:
Woofer choice depends a bit on box size, LROY fits in small box, XXLS10 or RSS225 might be nice too.

I recommend tower for L/R because more choices with drivers and easy to double the drivers to get lower distortion and lower tuning (or sealed). Subwoofer is not needed for music then, and hardly even for movies. But if you have good sub(s) already... and dsp channels to set nice slopes and handle delay.

I can see the benefits of of the tower, in particular with multiple drivers, but for now I'll stick to the standmount form factor. The plan doesn't include separate subs. I think, for example, stereo LROYs in 30l sealed enclosures will provide enough power and extension with the help of the room and dsp.
 
The difficulty of coaxials is how smoothly the tweeter sits inside the mid cone. Eg. MR18 and T18 from SEAS have pretty rough response aroud 8-12kHz because of this. SEAS C18 is better but the price is sky high! Simulations ignore these details!

Yes LROY sealed in 20-30 ltr box stereo gives nice bass! Very high spl is not it's bravure..



.
 

Attachments

  • ainogneo83 conf3+ 200220 spot.jpg
    ainogneo83 conf3+ 200220 spot.jpg
    186.3 KB · Views: 1,043
  • aingneo83 v7e disto 102dB inroom.jpg
    aingneo83 v7e disto 102dB inroom.jpg
    137.7 KB · Views: 230
Well a simulation tends to ignore details that were ignored when building the simulation:D Right now I'm just trying to learn how this kind of tweeter works in a waveguide. If I've understood correctly, the the surround makes a considerable part of the radiating structure, but it's not a real ring resonator because the dome can move freely.
 

Attachments

  • cox120.jpg
    cox120.jpg
    50.2 KB · Views: 325
  • tweet.jpg
    tweet.jpg
    43.6 KB · Views: 280
If you want coaxial 3way like Genelecs, then buy Genelecs, that should be said at first.

I mean competing with Genelec with retailed coaxial drivers is hardeous or even hopeless if I'm brutally honest.
Almost every single coaxial driver on the market has much worse off-axis behaviour typically in the top 1.5 Octave compared to Genelec's coaxial drivers and also tweeters with properly designed waveguides.
You only get vertical forgivingness of the listening point from these coaxial drivers in the exchange of good horizontal directivity, which doesn't sound fair at all in most applications to me.
Even if you add continuous flare, you may get better handling in lower frequencies, but you can hardly change these flaws.

Cheers.
 
I don't think I listed competing with Genelecs as a goal of this project and there's one less zero in my budget in the retail price of a 8260 pair. Other than that, I agree on most parts. The flare won't fix problems originating from the dome region. I'm not giving up yet though.
 
Sorry, I missed that part, your target was SP38/13, which is a bit of relief.
As long as I know, getting KEF SP1588 really was the best choice but second-hand LS50s (you're going to active so choose the cheaper previous model, optimally having a slight scratch on their enclosures) could be a pricy but very good option as well.
I think they have reasonably wide dispersion with a decently clean directivity, better than most retailed coaxial drivers.

If you can not get good coaxial drivers, and if you are not going to use them in the desktop/near-field, especially in the tight budget, I personally recommend conventional drivers with waveguides.
I mean, Neumann KH120s and KEF LS50s are roughly the same price range, both are respectable products. (personally prefer KHs)
And the important thing is though, if you are really good, you can do that level with conventional and not expensive drivers, but as long as you can't get very good coax, you virtually have no chance with coax.
(Still having greatest respect to your guts though)
 
Sorry, I missed that part, your target was SP38/13, which is a bit of relief.
As long as I know, getting KEF SP1588 really was the best choice but second-hand LS50s (you're going to active so choose the cheaper previous model, optimally having a slight scratch on their enclosures) could be a pricy but very good option as well.
I think they have reasonably wide dispersion with a decently clean directivity, better than most retailed coaxial drivers.

If you can not get good coaxial drivers, and if you are not going to use them in the desktop/near-field, especially in the tight budget, I personally recommend conventional drivers with waveguides.
I mean, Neumann KH120s and KEF LS50s are roughly the same price range, both are respectable products. (personally prefer KHs)
And the important thing is though, if you are really good, you can do that level with conventional and not expensive drivers, but as long as you can't get very good coax, you virtually have no chance with coax.
(Still having greatest respect to your guts though)

No worries. I appreciate your comments. I suppose I could have listed learning and having fun as the main goal (Edit: done), but I'll be dissapointed if I don't have a nice pair of speakers in the end. The driver choice is critical, since I don't think I have the experience required for overachieving with mediocre drivers. (I have more experience working with optical than acoustic waves). I still see Kef drivers as the safest way towards the coaxial direction. I'm not in a rush, so perhaps I'll entertain myself with the simulations while waiting for good deals.

It's interesting that you mentioned Kef SP1588. That's the 6.5" model, right? People seem to prefer Q100 (5.25") over Q300 (6.5"), but I suppose I should not judge those drivers based on how well they happen the perform in such low-cost product with super simple crossovers etc.
 
Last edited: