• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Question about coaxial drivers, directivity vs equalization

gorman

Active Member
Joined
Jun 26, 2020
Messages
268
Likes
167
Location
Italy
As per my other thread I have recently bought R7 Meta and R6 Meta speakers.

They are in two corners of an almost perfectly equilateral triangle with the last angle being the MLP. I currently have angled the R7s 15° toward the MLP, which still is significantly off axis. The R6 sits on a cabinet in front of the TV, below it, and is angled toward the MLP with an Aperta 300 isolation stand.

They are connected to a Denon X3800H, although the R7s are now driven by an Audiophonics MPA-S250NC XLR. I use Dirac DLBC, with three subwoofers (2 SVS SB1000 and 1 SB2000).

The question is this: I've used a slightly modified Harman curve in Dirac (I tame down sub 30Hz frequencies, in order to avoid resonances with the walls... since I've gotten the third sub they're better but I still live in an apartment, with people living below me). This is the curve I use. I use it for all speakers, according to Dirac recommendations.

22.6 0.0
25 3.0
30 6.6
40 6.4
50 6.2
60 5.9
70 5.5
80 5.2
90 4.8
100 4.5
120 3.5
140 2.5
170 1.5
190 0.9
200 0.6
220 0.5
240 0.4
300 0.3
400 0.2
18000 -3.5
20000 -4.5

The question is this: is there an effect derived from the directivity of coaxial drivers (the reason why it's suggested to have them off axis) that is not measured by a frequency response measurement in REW? Or if I tame the highs through Dirac I get more or less the same effect as I would get by moving them more off axis? I ask because currently they still feel a little bit too bright for my taste (nothing major, just want to get as close to perfection as I can).

Same question holds true for the R6 central speaker, which I currently point straight at the MLP (but I could point it lower through the stand's adjustments... it's currently as high as it can go and that makes it point precisely where my head is when sitting).

I hope the question is clear. First time with coaxial drivers I need to find my way with them.

(Most of all, I hope I'm using the directivity words correctly... but I could have totally misunderstood it so far. In that case, sorry in advance for my ignorance)
 
A coaxial speaker is going to project an even sound throughout the room, which makes the exact position of everything less important.
Directivity is how narrow a beam is. A laser is maximally directive. It has the highest directivity because it only goes in the exact direction the beam is pointing. A speaker with less directivity covers a broader area than exactly where it points. A perfect Omni speaker has no directivity.

What would be helpful is seeing a before and after graph from DIRAC, maybe more, like the correction curves. I can’t do anything with the data points provided. If you’re concerned about the measurement rig, a drawing or picture of the room would help as well.
 
What would be helpful is seeing a before and after graph from DIRAC, maybe more, like the correction curves. I can’t do anything with the data points provided. If you’re concerned about the measurement rig, a drawing or picture of the room would help as well.
I will come back with everything you ask. Thank you for answering! Much appreciated.
 
What would be helpful is seeing a before and after graph from DIRAC, maybe more, like the correction curves. I can’t do anything with the data points provided. If you’re concerned about the measurement rig, a drawing or picture of the room would help as well.
Room is 4.85m x 3.72m (15.9 x 12.2 feet), ceiling at 2.9m (9.5 feet) .
Ceiling and walls are acoustically treated with panels covering a considerable portion of the space (but not all, I did not want a "dead" room. I've also got three bass traps, placed in two corners and one wall (had to work around "normal life" concerns, much as this room is more or less dedicated).

This is how the front looks like:

1738271041771.png


The third subwoofer (SVS SB2000) is placed along the left wall, facing the sofa where I sit. Behind the sofa there's a bookshelf full of books. The illustrations you see on the wall are acoustic panels too. There's others on other walls.

Front speakers are 36cm / 14" from the back acoustic panels, measured from the end of the cabinet. They are toed in 15° but are well off axis from MLP still.
Center speaker is really close to back wall (but it's sealed, not ported), pretty much dead pointed to the MLP, completely on axis, as stated.
I could angle it down if that is desirable. I cannot aim it higher, though.

Distance from front speakers to MLP is 220cm / 7'2.6" (equal, Dirac "sees" it exactly at ), distance between the two front speakers is 230cm / 7'6.5" (measured tweeter to tweeter). Distance from centre speaker tweeter to MLP is 210cm / 6'10.6".

Target Curve
1738269897011.png


Front measured
1738269971949.png

Front corrected
1738270012434.png

Center measured

1738270049011.png


Center corrected
1738270104298.png
 
First a direct answer. Coaxials avoid by design a directivity error in the vertical, that most non-coaxials have. They may sound a bit more 'present' due to that, as in many implementation the 2kHz region is affected.

Second to that to get it 'precise' to a target is not the point. 'Perfection' is when you just like it. The target is a preference--is is an abstraction, because it is taken from many and then is averaged. You may want to find your personal 'sweet spot', simply what you like most. You have the advantage of an equalizer, measurement gear and very good material to begin with. The measurement gear will prevent you from running to far into the jungle of endless possibilities.

Don't worry, follow your feelings and that will be perfect.

Ehm, third, if you want to replicate the Harman anyway, do so. In case your adjustment looks a bit to hot in the region of 2kHz and up. It should be, as the average(d) listener would tell, a gently falling slope. Conversely, the region around 200Hz could benefit from a tad more level.
 
'Perfection' is when you just like it.
Oh, absolutely. That's how I meant it.

Generally speaking I've found Harman target serving me well. But this curve... Yeah, I'm now testing a slightly different one, with more of a straight slope down. Tomorrow I will test more and report back.

I just wanted to wrap my head around how coaxial drivers enter the overall equation.
 
Oh, absolutely. That's how I meant it.

Generally speaking I've found Harman target serving me well. But this curve... Yeah, I'm now testing a slightly different one, with more of a straight slope down. Tomorrow I will test more and report back.

I just wanted to wrap my head around how coaxial drivers enter the overall equation.
Also read this while "correcting" .
It comes from the one of the people who made the curve:

 
Also read this while "correcting" .
It comes from the one of the people who made the curve:

Well, one of the greats. Interesting read. I more or less agree with everything, luckily. A great fan of Auro3D, too.

Edit: you changed the post you linked to, as for the new one, well... I went with KEF as they seemed designed with the principles he enunciates firmly in mind and measurements seem to agree with that.
 
Last edited:
Well, one of the greats. Interesting read. I more or less agree with everything, luckily. A great fan of Auro3D, too. :)
Read again the new link,about correction *edited
Agree of course.
 
Coaxial drivers spread sound more evenly, so position isn’t as critical. But if it’s too bright, EQ tweaks help, though they’re not a full replacement for adjusting the angle. Trust your ears.
 
That looks awesome. The setup, the original measurements, and the corrections are all absolute minimum. To confirm that they are all working as intended, place the measurement mic at a few realistic listening positions. These will likely be different from the triangularly perfect center, but it will reveal speaker issues, placement issues and room issues which may need addressing. I can't see any right now, though. Your setup is phenomenal.

A good alternative is to take an average over time while moving the measurement microphone randomly through space. This will give a "room curve" that averages errors. In a complex environment, this can be helpful.
 
Agreed, what Dirac tells you it's going to do and what it does is usually two different things. MMM is excellent for evening out a listening area.
 
It looks like the area from 200hz-1khz is reduced after correction. That is not necessarily a good thing and might be perceived as brighter sounding. Do you find them to be bright without correction as well? Perhaps try to adjust the curve to more closely follow the natural tonal response of the speakers.

Also, sometimes "too much highs" is not really too much highs, but too little lows. So if "releasing" 200hz-1khz doesn't help, perhaps try to increase the low end with 1-2dB.
 
Thanks everybody for the inputs. I have modified the target curve a bit, following Heinrich's suggestions.
Highs were less bright but ever so slightly dull sounding, so I modified the curve to a sort of middle ground and it now sounds more "right.
Tomorrow, family needs permitting, I will do some MMM. Where REW will absolutely destroy the nice looking graphs Dirac offers (but that's expected and I don't care too much about the "smoothness" of graphs nowadays). It'll be interesting to see for sure.
 
Please share those REW graphs when you do them so others know the reality of Dirac predictions. In my case even DLBC has at times elevated midrange and treble regions even though I specifically wanted no corrections there. Don't be afraid to re-run Dirac, I've done it many times using anything from the wide listening area to the focused MLP and always end up with something slightly different but I've done it enough times to reject bad measurements and always correlate between REW sweeps and MMM.
 
Please share those REW graphs ...
Would you object if I say, DIRAC is lame? Just adjust to personal preference. The measurements are only to prevent from funny back and forth alterations. To find a preference spot is hard, really :)
 
You can say it but I would counter that to get the most out of it you have to put in some work and have a decent understanding of the information it presents you. It is not a simple "push the button and you're done" sort of correction. Regular Dirac is pretty useless to me as well but there's a bit of magic with DLBC in the <150Hz region that I haven't been able to replicate with my own filters.
 
Ok. I have taken measurements for L, R, L+R and C. Both with Dirac active and Dirac turned off.
The graphs have 1/48 smoothing applied.

L No correction
1738411535677.png


L Corrected
1738411574934.png


R No Correction
1738411616008.png

R Corrected
1738411644246.png

L+R No Correction
1738411673361.png

L+R Corrected
1738411698580.png

C Not Corrected
1738411722729.png

C Corrected
1738411743292.png

Target Curve
1738412055039.png

Let me know what you think, if there's anything that catches your attention, and so on and so forth.

I've attached the REW measurements file, in case you might find it of interest.
 

Attachments

  • MMM Dirac Harman Test.zip
    102.9 KB · Views: 17
@DWPress @Heinrich I take the liberty of tagging you because I don't know if you are receiving notifications for simple replies to this discussion.
Sorry in advance for the bother...
 
Back
Top Bottom