What?
Let me explain.
As I’m sure you know, the most popular speaker set in our industry is the 6-and-tweeter. Due to the lack of actual standards in our industry, sometimes it’s called a 6, sometimes it’s called a 6.25 or 6.5, sometimes it’s called a 6.75, sometimes a 165mm – but this 6 paired with a tweeter is the most common speaker set in our industry.
We have examples ranging from under a hundred bucks to literally thousands of dollars (American!) When it comes to component speakers, I estimate 75% of all units sold are 6-and-tweeter sets.
So, what’s wrong with them?
In the Educar classes, I talk about two principles of speaker operation which I have grandly dubbed the Law of Excursion and the Law of Dispersion – and each of these laws combine to wreck the performance of the 6-and-tweeter set in OEM locations.
First, let’s talk about the Law of Excursion. It tells that if you want a speaker to play an octave lower, and play at the same output level, it will have to move 4 times as far. (It doesn’t apply to vented enclosures, which minimize cone motion at the tuning frequency.)
So, let’s assume we are playing a 1k note – at a given output level, that requires cone motion of 0.1mm. Now, a tenth of a millimeter isn’t a cone motion you’re going to see with a six-inch speaker. You could put your finger on that cone and feel it hum, but if there was a logo on that dustup, you wouldn’t even see it blur.
An octave below 1000 Hz is 500 Hz. At 500 Hz we would be traveling 4/10 of a millimeter. Maybe that logo would blur at that point.
An octave below 500 is 250, and at 250 we would be moving 1.6mm. OK, at that point, we would definitely be able to see it moving, and it would probably be bouncing off our fingertip if we put it on the cone.
An octave below 250 is 125, and at 125 Hz, we would be traveling 6.4mm – which is roughly a quarter-inch of travel. That’s a lot of travel for a 6-inch speaker. A quarter-inch would probably be exceeding the safe excursion limits of the speaker. You’d also be in danger of hitting the back of the door panel with the surround in many installations.
But we haven’t even reached our usual crossover points yet. Let’s go one more octave, to 67.5, and see what happens. After all, many head-unit crossovers support a front high-pass of 65 Hz, right? At 67.5, we would be at 25.6mm of travel (that’s an inch, for us Yanks). There are a couple of 6-inch-class speakers with ridiculous levels of excursion, but none of them fit in the door of a Honda Civic.
This is a really cool exercise to do. When I first did it, I realized that asking for anything below 80 Hz was really a pipe dream on my part, and I started focusing on what Andy Wehmeyer had been telling me to focus on for years – the ramp up to the crossover point, and the phase interaction at the crossover point.
The test tracks on the CD that comes with the Audiofrog mic – similar to the tracks on the Educar TestTune app – let you play bass-only pink noise and evaluate the linearity of the midbass-to-subbass handoff really easily – so I now get the “bass up front” sensation really quickly and easily.
So, the Law of Excursion doesn’t limit us after all, as long as we use 80 Hz as our crossover point? Oh, it does – but mostly on the tweeters!
See, tweeters don’t have a lot of room for exclusion – and as many of us know from hard experience, play a tweeter too low and it rips itself apart.
Now, if you’ve seen my list of system design principles, you may remember that Rule #1 is “Don’t blow stuff up” (or words to that effect). When speaker manufacturers choose the crossover filter frequency for their tweeters, they take this principle to heart. They select a crossover filter frequency which makes tweeter failure in the field unlikely (as we know, we can’t make anything foolproof, because fools are so ingenious). The best way to keep a tweeter from blowing up is to use a high-enough crossover point and a steep filter slope. Well, when you design passive crossovers, every “order” adds a component. A first-order high-pass crossover is a series capacitor, and filters at a 6dB/octave rate. A second-order high-pass filter is a series cap and then a parallel indictor coil, and that filters at 12dB of attenuation per octave. Third order adds another series capacitor, after the parallel inductor, and gets us 18dB of attenuation per octave. A fourth-order gets us 24dB of attenuation per octave, and requires a second parallel inductor on the other side of the second series capacitor.
That means that steep passive crossover slopes cost more money (the mass of an air-core inductor is almost 100% copper). From a practical standpoint, most lower-cost speaker sets use a series capacitor as a first-order crossover filter on the tweeter, and better component sets use a second-order cap-and-coil, and that’s usually it. It’s very rare that we see a speaker manufacturer venture into a 3rd-order crossover filter for a tweeter, and 4th order is nearly unheard of.
This means that cost savings have conspired against allowing the tweeter to play too low.
And that’s where the Law of Dispersion comes in.
See, the Law of Dispersion says that as any piston plays higher and higher frequencies, its dispersion pattern narrows more and more, until its output comes straight out like a laser beam. “Beaming” as a principle says that every speaker has a range where it is basically omnidirectional in its dispersion, and above that point, it’s not. It gets worse and worse the higher you go. You can take steps to make it a bit better, but you can’t engineer this principle out of moving-piston speakers.
The Law of Dispersion starts to kick in around 1000 Hz for a 6.5” speaker, and it’s in full effect by 1500. By 2250, the output looks like a high-beam headlight on your car, and by 4500, it’s one of those flashlights my dad used to dial down all the way to a very tight beam for frog hunting. (As far as I know, he never went frog hunting, but that’s what he called them – frog-hunting flashlights).
This would not be a problem if the 6-inch-speakers were aimed at us, as they would be in a home speaker (or a demo board). But they aren’t – they are way off axis. This fact causes two problems.
Problem one is, the 6-inch-speakers in doors really roll off quite a bit in the vital 2000-4000 Hz range – there’s a lot of important midrange content in that octave.
Problem two is, the driver-side speaker is far more affected by this than the passenger=side speaker in most cars – because it’s much farther off axis relative to the driver’s seat. This exacerbates the Second Problem of Stereo quite a bit (One Side Sounds Different).
Why can’t we just drop the tweeter down farther, and get it playing those notes which the 6-inch struggles with? Because of the Law of Excursion. That tweeter better be pretty sturdy to be able to play low enough to matter.
This is one of the things that made the a/d/s/ 320i such a groundbreaking speaker in the early 80’s – in the age before neodymium was commercially usable, a/d/s/ made a tweeter (using samarium cobalt) that played down to 2500 Hz! Back then, tweeters from their competitors played to 6000 or 8000 Hz, so this was a huge improvement.
So, even today, some companies cross their tweeters over around 5000 or 6000 Hertz. They have their explanations for why they do that – but once they do, they are depending on the 6 to pick up the slack below the tweeter crossover point.
Now it is possible to make a speaker which has a rising response on-axis in the 1500-3000 Hz range, to offset this to some degree. The problem is, if the tweeter plays down to 2000 or so, its output at 2000 is symmetrical left or right. That is, due to the size of its piston, the tweeter has the same response in the driver’s door or in the passenger door. (Also, it sounds horrible in a display board if mounted at eye-and-ear level, which is where speakers sell best).
That 6 with a rising response, though, will still have a very different response in the driver’s door than it will in the passenger door. The only way to EQ that out is with asymmetrical EQ, which head units rarely have and which requires a DSP processor. Using EQ asymmetrically will help with a one-seat stereo car, but you don’t want to do that in a two-seat stereo car – it will just wreck your passenger-side response worse.
It’s better for a manufacturer to sell a speaker that stays out there intact, than it is to sell a speaker that sounds great in cars. It’s also more important to many speaker manufacturers who sell to large chain stores to make sure the speaker sounds great in the display board – even if that compromises the sound in a car. That’s because if it doesn’t sell itself in the display board, it won’t ever GET INSTALLED into the car.
Some companies – Hertz and Audison, Audiofrog, Morel, and Dynaudio come to mind – use crossovers in the 2k-3k range, and that helps offset this problem a good deal. It can make tweeters larger, harder to install, and definitely more costly, but it’s for a specific reason.
What’s the ultimate way around this? A 3-way front stage. That arrangement lets you operate each driver in its omnidirectional range, and get very similar response from each side. And that’s why 3-way speaker sets exist, and why many premium vehicles with premium sound systems use a front 3-way arrangement in their two-seat systems – because the small midrange is nearly omnidirectional within its range.
One of the reasons we sell the 6-and-tweeter set is because it fits into a lot of cars. Have you noticed it’s not as popular with the OEMs as it used to be? Now you know why!
