CSS Criton 3TD-X Kit Speaker Review

[amirm]

This is a review, listening tests and detailed measurements of the CSS Criton 3TD-X Kit tower speaker. It was sent to me by the company and costs US $2499 with a flatpack (pair). I received a fully built one which naturally costs a lot more:

Fit and finish is excellent, rivalling commercial speakers at this or higher costs. This is a 3-way configuration. The dual woofers should provide low distortion compared to bookshelf speakers with one.

Let's put it on the Klippel Near-field Scanner and see how it performs objectively.

CSS Audio Criton 3TD-X Speaker Measurements
Let's start with our family of anechoic frequency response measurements:

High level alignment of on-axis response is good. Zooming in we see some disturbances in midrange area and some boosting of treble above 10 kHz. We can figure out the sources of these in near-field measurements:

The resonances are tamed but still contribute slightly to on-axis response. Fortunately, off-axis is smoother:

As a result, predicted in-room response looks quite reasonable:

Speaker likely has a bit more "zing" which some folks may like. That is countered by deep bass reproduction (for its size) so in balance, it may sound fine.

Beamwidth is 20 degrees wider than average speakers I test so should project a wider, more diffused image (assuming you don't absorb side reflections):

Vertical directivity as usual is not as good but still allows some movement above tweeter axis:

The narrowing of the high frequency directivity likely counters some of the on-axis resonant peaking we saw earlier.

For distortion tests, I added a new range at 101 dBSPL:

As you see, response is quite good until we get to 101 dBSPL. Listening to that sweep, it still sounded pretty good. I initially tried 106 dBSPL and then there were some howls of discomfort from the speaker so the limit is somewhere between 101 and 106 dBSPL.

Impedance drops quite low so best to have a decent amplifier to drive it:

Waterfall display naturally shows the resonances we have seen:

Here is the step response for fans of that graph:

CSS Criton 3TD-X Speaker Listening Tests and EQ
In my large and reflective space, the 3TD-X filled the space with comfort producing deep bass and an "exciting" sound for lack of a better word. That extra excitement as I had predicted from measurements, came from slight boost in higher frequencies. I dialed those down and got to a more neutral stance:

I also dialed out a bit of that bass boost initially based on on-axis response. That was a mistake as speaker lost some of that excitement. So I followed the predicted-in-room response and got nicer results. That said, I am not sure in a blind AB test, whether someone would prefer the EQ or the stock sound.

I was impressed with the ability of the speaker to produce deep sub-bass. It attenuated it a bit and had a touch of distortion but perfectly serviceable and hugely better than any bookshelf speaker.

I then listened to my long list of reference tracks. There was not a single one that did not sound good! The sound was gorgeous, with clean bass and treble response. Spatial aspects were impressive, making you forget I was listening to just one speaker!

Conclusions
Objectively, the 3TD-X comes close to our target for frequency response and aces distortion measurements. A touch of EQ corrects former errors although some may not need it as out of box performance is still excellent. Subjective listening tests impressed me more than objective data, putting a smile on my face on track after track. I can easily say that this is the best KIT speaker I have tested.

I am happy to recommend the CSS Criton 3TD-X speaker.

------------
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/

 

[amirm]

Reserved for @AdamG to kindly post the specs.

Manufacturer Specifications:

SHOP | css-audio

www.css-audio.com

 

[Martin]

Interesting. The company offers upgraded capacitors in the tweeter circuit: Jantzen Superior Z-Cap for $500 and Jantzen Amber Z-Cap for $1,300.

It would be interesting to know if the company believes they produce a measurable difference and if they indeed do.

Martin

 

[amirm]

It would be interesting to know if the company believes they produce a measurable difference and if they indeed do.

They told me they don't make a measurable difference, either at component level or for the speaker as a whole.

 

[Rja4000]

Nice.
Not cheap for a kit speaker, but nice.
(I see the pair in Europe at 3k€)
Thanks for the review.

 

[pierre]

Would a simple EQ help this speaker? Answer is yes a bit: it lower the bridgthness, boosted the bass a bit and corrected lightly the midrange. It improve flatness on on-axis, listening window and PIR which is usually a good sign.

Score goes from 4.5 to 5.5. You can increase the score by having sharper (higher Q) EQ (see @malky76 post below).

 

[Maiky76]

Here is my take on the EQ.
Please report your findings, positive or negative!

For the score rational your journey starts here
Explanation for the sub score
The following EQs are “anechoic” EQs 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 is usually not optional… see hints there.

The raw data with corrected ER and PIR:

Score no EQ: 4.5
With Sub: 6.6

Spinorama with no EQ:

• Not very flat/smooth
• HF irregularities
• Some resonances
• 4-5k a bit hot?
• PIR overall slope a bit shallow?

Directivity:

Better stay at tweeter height or just above.
Horizontally, better toe-in the speakers by 10/15deg and have the axis crossing in front of the listening location, might help dosing the upper range.

EQ design:

I have generated two EQs. The APO config files are attached.

• The first one, labelled, LW is targeted at making the LW flat
• The second, labelled Score, starts with the first one and adds the score as an optimization variable.
• The EQs are designed in the context of regular stereo use i.e. domestic environment, no warranty is provided for a near field use in a studio environment although the LW might be better suited for this purpose.

Score EQ LW: 5.5
with sub: 7.3

Score EQ Score: 6.2
with sub: 8.1

CSS Criton 3TD-X APO EQ LW 96000Hz
November132024-144958

Preamp: -3.00 dB

Filter 1: ON HPQ Fc 40.3 Hz Gain 0.00 dB Q 1.39
Filter 2: ON PK Fc 98.4 Hz Gain -2.21 dB Q 1.19
Filter 3: ON PK Fc 587.1 Hz Gain -1.71 dB Q 3.16
Filter 4: ON PK Fc 1131.8 Hz Gain 1.41 dB Q 5.82
Filter 5: ON PK Fc 2234.1 Hz Gain 1.39 dB Q 0.90
Filter 6: ON PK Fc 9524.4 Hz Gain 2.43 dB Q 5.77
Filter 7: ON PK Fc 11396.5 Hz Gain -2.77 dB Q 5.75
Filter 8: ON PK Fc 14560.9 Hz Gain -2.60 dB Q 5.01



CSS Criton 3TD-X APO EQ Score 96000Hz
November132024-144958

Preamp: -3.00 dB

Filter 1: ON HPQ Fc 40.5 Hz Gain 0.00 dB Q 1.39
Filter 2: ON PK Fc 105.4 Hz Gain -2.21 dB Q 1.19
Filter 3: ON PK Fc 570.6 Hz Gain -1.37 dB Q 5.97
Filter 4: ON PK Fc 1103.3 Hz Gain 1.29 dB Q 5.93
Filter 5: ON PK Fc 2323.0 Hz Gain 1.36 dB Q 2.25
Filter 6: ON PK Fc 9469.9 Hz Gain 2.43 dB Q 5.94
Filter 7: ON PK Fc 11358.5 Hz Gain -2.83 dB Q 3.70
Filter 8: ON PK Fc 14555.9 Hz Gain -2.68 dB Q 5.81
Filter 9: ON PK Fc 4645.3 Hz Gain -1.54 dB Q 2.59

Spinorama EQ LW

Spinorama EQ Score

Zoom PIR-LW-ON

Regression - Tonal

Radar no EQ vs EQ score
Nice improvements

 

[Beave]

The high-Q wiggles from around 800Hz to 1kHz appear to be cabinet panel resonances, specifically the top and bottom panels.

If you look at the vertical off-axis plot (re-shown below), you can see narrow bands of energy radiating at just above 800Hz and just above 900Hz and centered on angles of 90 degrees above and 90 degrees below the on-axis (forward response). They go from about 50 degrees to 150 degrees for the top and -50 degrees to -150 degrees for the bottom panel. In other words, the narrow dips in the on-axis (forward response) at 800Hz and 900Hz are due to the top an bottom panels radiating the energy upwards and downwards.

 

[thewas]

Quite nice old school implementation which despite the lack of waveguide has not a big mid-tweeter directivity discontinuity, guess the smaller mid which probably also already reduces the radiation surface around the crossover frequency helps there.

 

[kemmler3D]

Looks decent for what it is, even for the money (which seems high for a kit?) seems credible all-around if you want higher SPL and a bit more bass out of your mains instead of the most linear response you can possibly get.

Still at this price I'd probably forgo the SPL and bass, stretch a bit, and get myself some Ascilabs or something like that, but I could see someone else wanting more oomph and going for this.

 

[Beave]

I just went back and edited my post to make it more clear what I'm referring to. I added some hand-drawn graphics to the plot.

 

[Ktacos]

The high-Q wiggles from around 800Hz to 1kHz appear to be cabinet panel resonances, specifically the top and bottom panels.

If you look at the vertical off-axis plot (re-shown below), you can see narrow bands of energy radiating at just above 800Hz and just above 900Hz and centered on angles of 90 degrees above and 90 degrees below the on-axis (forward response). They go from about 50 degrees to 150 degrees for the top and -50 degrees to -150 degrees for the bottom panel. In other words, the narrow dips in the on-axis (forward response) at 800Hz and 900Hz are due to the top an bottom panels radiating the energy upwards and downwards.


View attachment 405998

Prob not from cabinet resonances, more likely just diffraction from the tall baffle. Imp doesn't really show any issues there. With tall boxes the resonances associated with height are usually found much lower.

 

[Beave]

Prob not from cabinet resonances, more likely just diffraction from the tall baffle. Imp doesn't really show any issues there. With tall boxes the resonances associated with height are usually found much lower.

Possible. But the midrange driver seems much too far away from the bottom of the speaker for there to be diffraction issues at the bottom, yet there is energy radiating at 800 and 900Hz directly from the bottom of the speaker. The midrange driver isn't all that close to the top corner either.

 

[Beave]

For anybody who has built these or knows the internal construction, is the midrange driver in its own sealed compartment?

 

[JSmith]

It was sent to me by the company and costs US $2499

Thanks to them and for your testing Amir.

Their product page shows a 10% off voucher for an early BF sale and this is what the kit includes (standard crossover option);

A decent build video here.


JSmith

 

[tktran303]

out of box performance is still excellent. Subjective listening tests impressed me more than objective data, putting a smile on my face on track after track

Interesting.

IMHO-
Floorstanders, for the win!

 

[AnalogSteph]

The resonances are tamed but still contribute slightly to on-axis response.

Yeah, the woofers could do with a series notch, suppression around 1.5-1.8 kHz is decidedly less than the recommended 20 dB. (Probably a better investment than the fancy parts upgrade.) Not the most well-behaved drivers I've ever seen.

The ~2.7 ohm minimum strikes me as a bit low, too.

Overall though, pretty good.

 

[MZKM]

Reserved.

 

[BigSweeny]

It looks fairly nice in the finish that you got and great to see it performing well with the minimum EQ you did.
Personally it's a bit out of my budget for a DIY kit, however I can see the fun in it if someone had the cash and wanted a small project too.

Nice review and always good to see other options, thanks Amir!

 

[milosz]

Good build quality - decent quality parts in the crossover except maybe for that iron-core inductor in the woofer circuit - and all the inductors would be better if the wire was heavier gauge for lower loss; good stuffing, Wavecor bass and mid drivers. The cabinets look pretty nice, too.

I'd like to see a group delay graph. Vented speakers often have ugly bass group delay.