2. speakers
  3. 146621-38404

Small drivers vs big drivers

Hi,
I have a question that is always in my mind recently. I see some speakers with small  drivers (5-9 inches) that is reviewed to be able to throw out big sound stage and go down to 18hz-20hz. Some other speakers with big drivers (10-15 inches) though are commented to have 'big sound stage' but can only go as low as 30-35hz. 

To make the situation more complicated, some speakers have small drivers but there are many of them. Can many small drivers be compensate for the size limitation?

I don't know which specs determine a wide sound stage and the ability to reach low frequencies.  What is the pros and cons of each design?

Thank you!

Huy.
Ag insider logo xs@2xquanghuy147
audiokinesis
1,872 posts
 

Kosst, I think you are resorting to straw-man arguments; that is, putting words into the mouths of those you disagree with in order to make them appear foolish.

I design speakers having what you would probably call "giant drivers".

The primary reason we use them is for radiation pattern control, to minimize detrimental early reflections and create a spectrally-correct reverberant field.

One secondary benefit is improved dynamic contrast. Our giant drivers have negligible thermal compression at home audio SPL’s.

Another benefit is compatibility with a wider range of amplifiers.

Another benefit is a larger sweet spot with proper set-up.

Like just about everything else in audio, driver size is a juggling of tradeoffs.

Duke

kosst_amojan
2,095 posts
 
@audiokinesis 

I hate to break it to you, but you're wrong about radiation patterns. At every scale where waves are radiated and observed, arrays are the preferred and dominating technique. The largest radio telescopes are arrays. The largest optical telescopes are arrays. The largest audio rigs are arrays. The largest radio transmitters are arrays. 

Your argument concerned thermal effects is a non-starter because the basis of your argument assumes one large driver with one large voice coil vs. one smaller driver with one smaller voice coil. For an array of smaller drivers equalling the radiating area of one large driver the voice coil thermal radiating area will be greater in the smaller drivers. 

Ultimately the well understood physics of arrays better supports the technology. 
audiokinesis
1,872 posts
 

Kosst, of course we can compare arrays of smaller drivers vs large drivers.

"Array" is a rather open-ended term. How big is your array? Line array? J-array? Splayed array?

Arrays can get pretty big and pretty sophisticated.

So let’s do an actual apples-to-apples comparison.

Each of us picks out about $300 worth of drivers for their hypothetical speaker. You specify the shape of your array. Then let’s look at thermal and mechanical limits, radiation patterns, and bandwidth.

I’m picking a $210 woofer (Eminence Kappalite 3012LF), a $45 compression driver (Celestion CDX1-1446), and a $14 horn (Dayton Audio H6512), Parts Express retail prices. That comes to $269. These are all components that I actually use.

You pick your ballpark $300 worth of array drivers, tell me how they arrayed, and let’s take a look at the thermal and mechanical limits, radiation patterns, and bandwidths. I have a modeling program that can model radiation patterns fairly accurately, "Enclosure Shop" by LinearX. I will post my analysis of how your array would perform, and of how my giant driver two-way would perform. You are of course invited to do your own analysis.

Let’s get specific. Maybe you are right, maybe your array will be better across the board. Let’s go apples-to-apples and find out.

Duke

audiokinesis
1,872 posts
 

Kosst: "I hate to break it to you, but you’re wrong about radiation patterns."

That’s possible. I’m not infallible.

Would you mind pointing out exactly what I said wrong about radiation patterns? I think it’s fair for me to ask that so I can take another look at whatever it was, especially given your tone of condescending certainty.

Thanks.

Duke

kosst_amojan
2,095 posts
 
You weren't exactly specific about radiation patterns, but it sounded like you're suggesting a large driver has a superior radiation pattern to a smaller one. I'm not sure how that makes any difference unless you're trying to drive something like a 12" driver up over 1000Hz. That seems high in a 3 way speaker. 500Hz seems on the high side of reasonable and at that point a 12" driver isn't coupling to the air that much better than a 6.5" driver. 

In my bedroom I have an old pair of ESS AMT1s with 10" drivers. The area of that cone is pretty close to the area of the 3 6.5" drivers in my Focal 936"s. The bass from the ESS is substantially inferior to the Focal. Likewise, when I bought my 936's I listened only to the 926 and 948 in the same room right next to each other. If the presumptions made about large drivers and large radiating areas being that much better we're true, I can't think of a better circumstance in which to observe it. Naturally the 928 kinda fell apart below 40Hz, but north of that there wasn't any lack of dynamics or power aside from the 948 being a little more sensitive. Beyond that, the 2 6.5" woofers in the Kanta 2 didn't pan out to sound or measure significantly inferior to speakers with larger drivers.

If my understanding of line array theory is right, the stack of 3 drivers in the Focals has the vertical directivity of a 20" woofer because of how they mechanically couple. Not that it's really significant since the start rolling off at 150Hz and are crossed over to the mid at 260Hz making their pass band wavelength much larger than 20". Even at 500Hz the wavelength would be a good bit larger than the cone of a 20" driver rendering it primarily omnidirectional. 

I'm certainly not saying an array of drivers is cheaper than one big one, but some people out there are suggesting that. In my opinion it's a question of which technical challenges one wishes to take on. Making really stiff, light, well damped, large cones is a real challenge of one sort. Effectively integrating multiple drivers into a cabinet and keeping it price competitive is another kind of challenge. I don't find that dispersion or thermal effects to be real issues unless we're going to make the lopsided comparison of oranges to watermelons. In terms of area of cone moved, 3 smaller drivers will shed more heat and be more accurate than a 10" driver. It's the more expensive solution to the problem.

As for specific drivers, I wouldn't even know where to begin. I could only guess at what each of the 6.5" F cones in these Focals would cost if they sold them. $150-200 each? They sell the 10" F cone for $350 each. That kind of thing would blow your budget fast. 


More to discover