Imaging: coaxial vs. traditional layout

[RobL]

I have a bit of spare time today and thought I’d do a little comparison of my 8361A’s vs. my 705P’s, at lowish levels and about 4’ and 10’ listening distances.
It’s been mentioned here that coaxial drivers lose their advantage (re: immediate integration) outside of near field conditions. When I first listened to my ‘61s the first thing that came to mind was “that’s an unfamiliar presentation!”. I just couldn’t put my finger on it, other than the soundstage seemed a little narrower than I was used too. I thought a direct comparison might reveal any other differences to me.
There actually are other differences, even in far-field! Using the track “Hey Brother” by Avicii as an example…The song begins with alternating guitar notes for several seconds before the singer comes in. On the 705P’s the notes are almost amorphous, seeming to (almost?) alternate from left to right at the outside of the soundstage. The effect is consistent from either distance. With the 61’s, the guitar is a solid image placed into the soundstage just left of where the singer appears. It’s also consistent at either distance. Both monitors give a solid image of the singer in the centre of the stage.
Which presentation is correct? Why do the two near field monitors give very different images with the same song from the same location? Has anyone else noticed similar effects?

 

[fpitas]

I went to MTM precisely to stop the image from moving as the sound transfers from the midrange drivers to the horn. I think a non-moving image is "right". Real music doesn't wiggle around because the frequency of a note rises and falls.

 

[KSTR]

There actually are other differences, even in far-field! Using the track “Hey Brother” by Avicii as an example…The song begins with alternating guitar notes for several seconds before the singer comes in. On the 705P’s the notes are almost amorphous, seeming to (almost?) alternate from left to right at the outside of the soundstage. The effect is consistent from either distance. With the 61’s, the guitar is a solid image placed into the soundstage just left of where the singer appears. It’s also consistent at either distance. Both monitors give a solid image of the singer in the centre of the stage.
Which presentation is correct? Why do the two near field monitors give very different images with the same song from the same location? Has anyone else noticed similar effects?

I had a very similar experience almost 20 years back when I switched from Genelec 1031A to Tannoy System600A. The 600A is by no means a flawless monitor but its imaging capabilities got me hooked. Hooked again, that is, because years before I spent quite some time with Tannoy 15" DMT-II large coaxials, an encounter that still has impact on me, even today ;-]
Image stability and precision is not a given for a coaxial, the exact XO design, diffraction, resonances etc will all play their part. But when it happens, it tends to be really convincing. I would attribute that to the better correlation of off-axis to on-axis sound (and, designing for direct on-axis is problematic with any coax IME).

 

[RobL]

I don’t think it’s a switch of drivers though…could it be a phase thing? The 8361 is basically phase linear from 800hz up, and the 705p goes through three 365 degree phase rotations up to 20khz according to S&R. Does a coaxial have a phase advantage over conventional layout?

 

[fpitas]

I don’t think it’s a switch of drivers though…could it be a phase thing? The 8361 is basically phase linear from 800hz up, and the 705p goes through three 365 degree phase rotations up to 20khz according to S&R. Does a coaxial have a phase advantage over conventional layout?

In my case the crossover was LR4, in-phase, yet the image noticeably moved vertically at times.

 

[KSTR]

Does a coaxial have a phase advantage over conventional layout?

Well, in many coaxials the tweeter is (way) behind the midrange or woofer cone which allows for more flexibility in crossover design, including quasi phase-linear crossovers, sometimes. Having group delay headroom on the woofer is always a good thing and much better than when the tweeter is leading the woofer.

 

[Sancus]

The Ones are phase linear from 100hz up if you turn extended phase linearity on, and it only adds like 4ms of latency, so I just keep it on at all times.

 

[RobL]

I think I found the cause, I seem to have a defective cable.
Apologies! Nothing to see here it seems…I’ll keep listening lol

 

[MKR]

I think I found the cause, I seem to have a defective cable.
Apologies! Nothing to see here it seems…I’ll keep listening lol

Then the conclusion is? Coaxial sounds same near and far field vs traditional? I had also always understood what you stated to be true ... coaxial advantages are only apparent in the near field, but become less so for mid/far-field MLP

 

[RobL]

Then the conclusion is? Coaxial sounds same near and far field vs traditional? I had also always understood what you stated to be true ... coaxial advantages are only apparent in the near field, but become less so for mid/far-field MLP

Well, I’m going to rearrange things here a bit and try again. I’ll post whatever I can scrutinize when I get them on equal footing

 

[RobL]

My first interesting discovery. My DAC’s sound different
Seriously, my airport express’s (I have 3) all sound quite different from my RME Adi2.
I wrongly assumed I could run analog out from the RME to one speaker and analog from an AE to the other speaker so I could quick-switch for this with differences being small enough as to be inconsequential. Nope. The AE is causing the imaging difference, not the cable. I swapped speakers and AE’s to see if it was a one-off. No, they all image the same and quite differently from the RME.
What the hell?

 

[McFly]

Seriously, my airport express’s

What the hell?

 

[RobL]

Ken Rockwell measured the audio output of them HERE.
Where do you think they’re defective?
My first thought was it’s an airplay issue, but the RME is also fed a digital signal from an AE over airplay. They are all also on firmware 7.6.9

 

[RobL]

Is there any chance somone has an Airport Express they can connect to some monitors to verify this?
Use the “Hey Brother” track, the imaging difference isn’t subtle.

 

[Sangram]

Well, in many coaxials the tweeter is (way) behind the midrange or woofer cone.

In most concentrics the woofer and tweeter are at the same Z-distance from the listener. The point of origin is assumed to be the junction of the coil and cone, and not the cone itself. This allows time alignment of both drivers, and opens up a few more possibilities of crossover design, while causing termination errors at the physical transition point between them. These errors usually improve off-axis, thus the overall listening setup needs to be a bit different from non-coincident layouts.

 

[AnalogSteph]

I see the Airport Express has a 3.5 mm output. Certain combinations of plugs and jacks can be prone to ground contact issues, so that may be worth looking into. A crosstalk measurement would show any such issues clear as day. You can try e.g. REW's signal generator to see whether L+R is severely attenuated over L-R (L+R with second output inverted) - best turn on only one speaker at a time for this test.

 

[RobL]

I see the Airport Express has a 3.5 mm output. Certain combinations of plugs and jacks can be prone to ground contact issues, so that may be worth looking into. A crosstalk measurement would show any such issues clear as day. You can try e.g. REW's signal generator to see whether L+R is severely attenuated over L-R (L+R with second output inverted) - best turn on only one speaker at a time for this test.

Thanks, I might try that, though I get the same result from all three AE’s and even my amazon Alexa (from the 3.5mm out). Using different cables didn’t change anything. This is driving me a bit nuts. From the RME I get a defined location of the guitar, from all the others an undefined diffuse sound. All the other images are correct between all the units.
Can anyone else tell me what they hear on their systems, I’m not sure which is the correct presentation now.

 

[Juhazi]

In most concentrics the woofer and tweeter are at the same Z-distance from the listener. The point of origin is assumed to be the junction of the coil and cone, and not the cone itself. This allows time alignment of both drivers, and opens up a few more possibilities of crossover design, while causing termination errors at the physical transition point between them. These errors usually improve off-axis, thus the overall listening setup needs to be a bit different from non-coincident layouts.

No, that is a myth that originates from PA horn+woofer constructions eg. old Altecs, "a rule of thumb" and era before FFT analysis of audio signal. Origin is at driver membrane that makes air molecules vibrate.

Typical domestic speakers are "direct radiating". This is easily verified with modern 2-channel acoustic measurements, as time delay difference. Even compression drivers in deep horns have origin at membrane

Loudspeaker time alignment - Wikipedia

en.wikipedia.org

 

[KSTR]

Origin is at diver membrane that makes air molecules vibrate.

This!
For a large mid-woofer with a steep cone for bass and midrange it means the acoustic center is quite a bit in front, many centimeters actually, of the VC/former junction, let alone the VC itself. In the treble when the bigger part of the membrane (hopefully) gets damped and decoupled from the point-of-entry area the acoustic center shifts back a bit, though. But still not to close to the VC/former joint.

I'm mostly interested in PA or higher power classic coaxials and all of these have compression drivers in HF section, with their membrane again many centimeters behind the woofer's acoustic center. I'm aware that there are a few HiFi/Monitor concentrics with dome tweeters which are flush-mounted within the midrange membrane.

 

[outfaced]

The Ones are phase linear from 100hz up if you turn extended phase linearity on, and it only adds like 4ms of latency, so I just keep it on at all times.

for me the question is, do you here the difference with on and off and what is like ... more like just different, or really better in any regard when on?