Omnes audio cx3.3 coaxial or SB 15nbac+ SB 26adc ?

[Tangband]

The Kef uni-Q drivers seems very interesting as a point source but they are expensive .

The Omnes cx3.3 coaxial driver seems to be an interesting cheaper alternative and all reviews I have read so far is positive , the price is only 45 euro. The distortion seems to be low. This driver seems suitable for making a LS 60 clone on a budget.

As you all know, the SB nbac15 used as a midrange with SB 26adc seems to be one of the best technical solutions between 250 - 20000 Hz , but I wonder if this omnes cx3.3 is maybe an even better option between 250 - 20000 Hz having the benefit (?) of being a coaxial driver ?

1. Is has a 4 inch metal midrange cone .

2. It has a 0,75 inch metal dome tweeter .

The driver is here :

Omnes Audio CX 3.3 - loudspeaker diy, 55,00 €

diy loudspeaker Point Source Speaker

oaudio.de

I want your opinion about the best solution between 250-20000 Hz in DIY of a dsp threeway.

 

[SomeDude]

For the price it's a very good driver. There was a good review with a lot of measurements on the German site: "hifi-selbstbau.de" but unfortunately it seems to be no longer for free...
The only real weakness of this driver is it's power handling / SPL capability and it is recommended to cross it at 400 hz if you want to use it in a three-way.
Take a look at the Dreizwo

Best solution for a three-way for 250-20 khz? That depends on too many criteria: How loud, how big and what directivity?

 

[thewas]

Its directivity of course cannot be compared to the KEF drivers and it needs some care at the lowpass of the woofer

(source and more: https://www.hifitest.de/test/lautsprecherchassis-breitbaender/omnes-audio-cx33-21855)

but for its low price absolutely decent.

The SB drivers will outperform them in terms of non-linearities but also might have other directivity issues so the choice depends more on the overall tuning and usage, for example for nearfield usage a coaxial design can have significant advantages.

 

[Tangband]

Its directivity of course cannot be compared to the KEF drivers and it needs some care at the lowpass of the woofer

(source and more: https://www.hifitest.de/test/lautsprecherchassis-breitbaender/omnes-audio-cx33-21855)

but for its low price absolutely decent.

The SB drivers will outperform them in terms of non-linearities but also might have other directivity issues so the choice depends more on the overall tuning and usage, for example for nearfield usage a coaxial design can have significant advantages.

Thanks for input. Really low distortion though, especially for the money . The tweeter decay are also clean . The midrange seems to have some unregularities but this can probably be dealt with , with a dsp crossover and some peq.
The distortion figure are at 85 dB.

 

[Tangband]

Thewas : a crossover 24 dB/oct at 3500 kHz seems to give rather good directivity though ? Its about the same directivity for the midrange and tweeter at that frequency ?

However - This driver might be a nightmare with passive crossover, because the peak between 1800 Hz - 4 kHz , +4 dB in the midrange must be dealt with ( can be done with dsp peq ). The breakup peak at 6,5 kHz for the midrange demands a steep crossover, if its gonna be at 3500 Hz. A correction notch crossover for 6,5 kHz could be done passive I guess, but there is still the peak 1800 Hz - 4kHz to deal with.

It would be nice If someone could simulate a passive crossover for this driver.

 

[Tangband]

For the price it's a very good driver. There was a good review with a lot of measurements on the German site: "hifi-selbstbau.de" but unfortunately it seems to be no longer for free...
The only real weakness of this driver is it's power handling / SPL capability and it is recommended to cross it at 400 hz if you want to use it in a three-way.
Take a look at the Dreizwo

Best solution for a three-way for 250-20 khz? That depends on too many criteria: How loud, how big and what directivity?

Thanks for the building link .

 

[Tangband]

Omnes Audio CX3.3

Test Lautsprecherchassis Breitbänder 2024: Unsere Redaktion hat den Omnes Audio CX3.3 für Sie unter die Lupe genommen

www.hifitest.de

”With the small membrane of the CX 3.3, however, this is not a problem. The rubber surround is remarkably narrow, indicating that it can be used from 200 Hz upwards. With the high resonant frequency, the small coax doesn't even try to play bass, but focuses on what it's good at: midrange and treble.

In the center of the aluminum cone, behind a fine-meshed grille and a large-area diffuser, is the corresponding high-frequency counterpart, surrounded by a fabric bead. Ventilation openings in the basket of the mid-range part are unnecessary in view of the small cone movements, but due to the small basket openings, care should be taken to ensure an “airy” baffle cut-out to the rear.
Zeroing in on pure midrange gives the CX 3.3 advantages in terms of efficiency: a full 88 dB nominal level is available from 300 Hz. Apart from harmless fluctuations between 700 Hz and 2 kHz, the frequency response is very linear and surprisingly good-natured up to over 5 kHz, where unfortunately a strong resonance can be found - the connection to the tweeter shouldn't be a problem. The distortion also gives a green light above the fundamental tone, and the compact coax can easily handle even 95 dB. The decay is very clean and fast, and almost perfectly error-free despite the aluminum membrane.

The treble part is just as easy. From 2 kHz full level, falling slightly above that, the frequency response can be smoothed out from above without any problems. Slight irregularities are averaged out under angles, the coax-typical, narrow-band dips are completely absent. The resonant frequency just under 2 kHz and the very low distortion from 3 kHz suggest its use from this frequency; The user is on the absolutely safe side and still easily feasible with 3.5 kHz - the manufacturer's recommendation. Here, too, the decay behavior is very clean, fast and also free of resonance.

The parameters refer unmistakably to the use from the lower midrange in three closed liters.”

 

[SomeDude]

Thanks for the building link .

I made a mistake that was the link to the old version...
There is a newer version.

 

[Tangband]

I made a mistake that was the link to the old version...
There is a newer version.

Thanks, very interesting !

 

[thewas]

Thewas : a crossover 24 dB/oct at 3500 kHz seems to give rather good directivity though ? Its about the same directivity for the midrange and tweeter at that frequency ?

I was talking more about the directivity and edge diffraction issues of the drivers themselves as the directivity step between the two drivers can be minimised by choosing the appropriate crossover frequency and possibly also not too steep slopes.

 

[Tangband]

I was talking more about the directivity and edge diffraction issues of the drivers themselves as the directivity step between the two drivers can be minimised by choosing the appropriate crossover frequency and possibly also not too steep slopes.

Aha, maybe a 12 dB/oct crossover with a passive notch filter then, for the midrange ?
There is some diffraction I guess coming from the rubber edges of the cone.

 

[thewas]

Also from the edges around the tweeter to the transition to the woofer, only very few manufacturers have perfectioned those. The optimal crossover has to be find by measuring and simulating various configurations.

 

[Tangband]

Another beautifully made build DIY with Omnes cx3.1

 

[Tangband]

Here is a building of a monitor speaker with passive crossover schema for Omnes cx3.1,
Comments on this ?

 

[fluid]

I want your opinion about the best solution between 250-20000 Hz in DIY of a dsp threeway.

A fairly close DIY contender to look at is also the Sica coaxes that come in a 5.5" or 6.5" Chassis. Apart from a dip at 10K (which is difficult to avoid for a coax that is allowed to have any real excursion) it is very well behaved.

Here is a Vituix sim I made of the 5.5" based on someone else's data, linked below. It is not minimalist on the EQ
https://www.diyaudio.com/community/threads/a-3-way-design-study.376620/post-7107630