Warm sounding and engaging full range speaker

[TEK]

I was a bit inspiered by this thread

https://www.htguide.com/forum/showthread.php?42802-Scan-Speak-D2904-7100-vs-D3004-6600

Where the basic point as I read it was that you can't just compare two tweeters.
I have no problem relating to that.

But I'm a bit curious.
If I was to say that I wanted a "warm" sounding and engaging speaker for a living room setup in the higher end of the quality scale. The focus would be on being engaged by the music and having a pleasing sound (for example - no shart s-sounds), and not necessarily getting all minor details perfect.
Limitations would be that it should be "normal" sizes (less than 150cm height, 40cm wide and 60cm deep).

Is that something that could be considered in the design phase?
How would you address such a question?
What factors would you focus on?

[JonMarsh]

TEK, as I see it, this is something that can certainly be addressed; it's a function of making specific voicing choices in the speaker design, while still keeping other linear and nonlinear distortion at a reasonably low level, considering the limitations imposed by size, Hoffman's iron law, available drivers, etc.

As regards voicing, what you'd be looking at is not flat, but with a controlled downward tilt, and possibly also a BBC dip. Sigfried Linkwitz came around to the necessity for this voicing with his Orion system, and in the case of a dipole, I think it's particularly relevant, because the power response is so uniform with regards to frequency, where as a conventional monopole radiator intrinsically will have some degree of tapered power response due to dispersion limitations of the presence range and treble, compared with the lower midrange and bass. We discussed the possible desirability of this voicing and Sigfried's published info somewhere in the three way design study, and I think in the first Ardent thread.

A commercial speaker line which follows this philosophy is Peak Consult, from Denmark.

This would be an example of some voicing along those lines for an MTM-



This response curve for the Peak Consult El Diablo shows taking that concept further:

[TEK]

I notice that the "warm" curve that you describe here is simular to the "Audyssey reference curve". That is the curve that the Audyssey room correction system often have setup as the default target when doing room correction.
They claim it is because the way movies are mixed compared to the normal listening environment.
That curve is not nessesarly recommended for music as music is mixed differently - they claim.

In the previous post Jon mainly claims that the tool to get a "warm" sound is by modifying the frequency curve.
Is that the only factor worth looking at? What about different type of membrane material? How would that affect the sound?
What about the crossover point's? Would it be preferrable to get the crossover point at a higher or lover frequenzy - that would probably affect what drivers you would choose?

[Face]

In the previous post Jon mainly claims that the tool to get a "warm" sound is by modifying the frequency curve.
Is that the only factor worth looking at? What about different type of membrane material? How would that affect the sound?
What about the crossover point's? Would it be preferrable to get the crossover point at a higher or lover frequenzy - that would probably affect what drivers you would choose?

It's more FR than anything else you mentioned.

[Carl V]

I'll throw something out...it's what I think of as the Power region of Music.
Bass weight & this isn't bass extension. Think of 80Hz to lets say 580Hz.
the exact numbers aren't critical. so,100-500Hz is a good starting point.
And in room Power response is critical... Not just freq. response graphs
but good Polar response/intensity graphs.

You won't get there with a nice 150mm midwoofer augmented with a 12 sub. imho.

http://www.independentrecording.net/irn/resources/freqchart/main_display.htm

I am confident most of us have seen this chart or something similar.

but look again. I just heard some speakers recently which all used larger
diaphragm drivers. AE, Beyma SEAS..etc., SD & VAS numbers were good.
Fun musical reproduction

[JonMarsh]

In the previous post Jon mainly claims that the tool to get a "warm" sound is by modifying the frequency curve.
Is that the only factor worth looking at? What about different type of membrane material? How would that affect the sound?
What about the crossover point's? Would it be preferrable to get the crossover point at a higher or lover frequenzy - that would probably affect what drivers you would choose?

One of the sometimes overlooked things about frequency response is linear distortion that is audible but not obvious in making a frequency response sweep. For example, if you have a membrane material with a high resonance break up mode, and don't suppress it sufficiently (a common issue with designs with Seas Magnesium drivers or Dayton RS Aluminum drivers) you can have an edge in the sound that shouldn't be there, that is akin to nonlinear distortion, but isn't exactly that- though resonance amplification CAN increase the level motor distortion products. With the right design, there's no problem in getting a "warm" sounding RS180 based system, for example.

All of the other things you mention are just the kind of things that have to be done right for any system to really sound good- whether it's voiced neutrally, forwards, or warm and recessed.

BTW, a drawback to pleasantly warm systems is that they often done't image quite as well, unless they're a dipole speaker like a late model Orion or a Quad ESL. We use a lot of the presence range and high frequency range stuff for imaging cues; getting good imaging is about suppressing spurious reflections in that frequency range and other areas, and reproducing the subtle room acoustic cues clearly of the original recording.

I've found this issue to extend to electronics- for example, my Marantz SACD players, while highly regarded as warm and musical, and having a pleasant playback tonality with a variety of recordings, don't image anywhere near as well as my Berkely Audio Alpha DAC. And for those for whom that comparison is a just a little too Audiophile, well, I have to call it like I hear it. :W

[Scottg]

Is that something that could be considered in the design phase?
How would you address such a question?
What factors would you focus on?

Yup. You can and should consider it in the design phase.

A rather large range of issues "pop-up" - which at this point (for me) becomes an almost intuitive/automatic response.


1. Your cabinet size provides some limits with respect to baffle support and loss along with interior volume and potential bass extension near the average for a particular bass driver(s).

2. "Warm" is as already noted.. though you can extend this further by increasing pressure loss at higher freq.s and increasing pressure gain at lower freq.s.. "Warmth" however is most generally a lower midrange attribute - so basically boost your baffle-step loss by almost the full +6db (where ordinarily you might only do so by +3 to +4 db depending on size of the baffle and the location of the driver relative to the floor). "Warmth" can be generated via a higher driver "Q" and/or a smaller volume enclosure for the given driver. "Warmth" can also be generated via the cone material and its modal suppression near the surround - a bit of "flex" and then some damping. Larger diameter drivers naturally flex more, as can drivers with membranes that flex at their edges - particularly poly, as well as paper, and even some of the poly-based "hex" cones (..that don't flex a great deal anywhere but their edges). Finally, "warmth" can be added with various component-types in a passive crossover - oil-damped poly capacitors often add to this a bit as can some of the doping on resistors.

3. "Higher end" is generally about depth perception. The fact is that you can spend very little money on a "cheap" speaker and get a very linear response - even a reasonably linear polar response. You can even do so with a diffraction signature that isn't much different than very expensive loudspeakers. The key factor in the better selling "Hi End" speakers is the ability to portray depth, and even give some "body" to "images". In large measure depth is a linear excursion and low-level quality dealing with the driver's suspension and in-box compliance along with the surround and how little it damps the driver's modal character (..its diaphragm). Look carefully at the driver's mechanical loss - it should be low. The surround in particular should be of as low a mass as possible (..and just because the material is normally "low mass" doesn't mean that the end result is - be careful.) Then load the driver in a manner that's going to provide very little resistance (low friction). Note that driver + loaded air compliance is a bit different than simply the driver alone - the larger the volume, generally the lower the level of resistance at very low excursion potentials (provided the spider and surround has been "broken in"). For drivers like tweeters with their own "set" rear chambers, look for the same sort of details: larger volume, relatively unobstructed air-flow with fairly low damping material in an around the rear chambers (..which is different than damping that's part of the diaphragm), and finally surrounds that are very light-weight and not overly damped with compound ("doped"). Be wary of components and added resistance in the crossover - particularly for the tweeter.

4. For avoiding sibilance or a "sharp" or "thin" character - even a "ghosted" "image", generally look to drivers with greater damping and more mass - but "flexing" mass, not exceedingly rigid mass - at least not near the surround. Because most sibilance resides at high freq.s: truly soft domes are a good start, but also well-damped ribbons (the non-pleated kind like the Neo series from Fountek). But it should also be taken lower in freq. as well - at least down to the lower midrange. Finally, consider diffraction as sound-waves "move" around the baffle - the material being used can effect the sound both from the surface quality of the material and any resonant issues. Plastic (low density variety) tends to sound like "plastic" (as if rubbing your finger across it and hearing the higher pitched "white noise" sound, and thin aluminum often sounds similar). A properly absorptive fabric layer can help a lot.. but it can also run a bit "afoul" of an added friction element mentioned in #3 (particularly near the tweeter).


..a LOT of other stuff to consider as well, but that should be more than enough as a "starter" for some of the more specific issues asked about.

[---k---]

I'm going to add, that I've been reading these forums for probably 10 years now, built a few speakers, listened to a few more. I see FR graphs posted all the time. To the casual observer I don't think a lot can be inferred. Jon, Matt, CJD and a few others can probably extract more information out of them, but, not me. Maybe I need to take a class...

For example, I've heard both of these speakers in the same room:


vs



I would say one sounded detailed, crisp, maybe slightly forwarded and lean bass. Music was presented in 3-d, but it seemed between and in front of the speakers. The other sounded rich, warm, laid-back. Deeper, more feel-it bass. It has more detail, but you had to listen more carefully to extract it. Music was presented in 3-d, but it was between and behind the speakers.

The graphs don't model the bass. and the scales are different. Anyone want to guess which is which? Don't cheat!