Originally Posted by mitchyz250f View Post
This is way over my head. But I think I am following most of it.
It still gives me headaches, but I think it's essential to understand it if we have any hope of creating a decent soundstage. When I see guys putting eights and tens in their car doors, it makes me cringe, because it doesn't work. (Been there, done that.)
The key to all of this is to understand that there's a transition between our perception around 1.6khz. In the one octave between 1khz we have to pay attention to both pathlength and frequency response. But below this octave pathlength is king, and above this octave frequency response rules. This transition is due to the distance between our ears, and their shape. If you want to, get a ruler and measure the distance. You'll see that the dimensions correspond to the frequency. (1600hz is 8.4".)
Originally Posted by mitchyz250f View Post
It seems that most of the benefit could be achieved with two tweeters at the center of the dash as Chithead experimented with and the picture dlechner posted (post 7)? These two tweeters would be off axis the same amount for both passendger and driver. Equalization and TA would be nearly identical for both tweeters.
Wouldn't that eliminate much of the need for additional data processing such as steering?
Would our brains identify the similar sounds from both tweeters as mono?
Honestly, I am not convinced that we need two tweeters at this point. Above a certain frequency, we're barely able to perceive depth or width. I am trying to figure out what the downsides to a single mono tweeter are, and I'm drawing a blank.
Here's the best I can come up with:
* If you use one tweeter instead of two, you must limit it's response to a range where the influence of interaural time delay is inaudible. According to Blauert, this happens above 1600hz. Because our tweeter will have output below the crossover frequency, it would be advisable to increase the crossover point to a frequency above the perceptual "crossover." Perhaps 3200hz, one octave above?
* Because interaural intensity differences are critical to soundstaging above 1600hz, if you DO use two tweeters, they MUST match. There are two solutions to this. Either put them both in front of you, where their directivity is similar. Or if you mount them in a different location, juggle their distance, directivity and frequency response so that you end up with the same goal: identical frequency response on the left and the right.
The thing is, it's a helluva lot harder to get two tweeters to sound identical when you're listening to one of them off axis, and their pathlengths are significantly different.
On the contrary, here are the drawbacks of stereo tweeters:
* Due to the Haas Effect, the tweeter on the opposite side of the car needs to be about 3-6dB louder. Compounding the problem is the fact that we're already too close to the driver's side tweeter.
* We are trying to create a soundstage in front of us. Due to head related transfer functions, when we move tweeters to the door or the kick panels it creates the perception that the sound is in the wrong location. It *is* possible to offset this perception via EQ, but I've never seen anyone publish a good technical article on how to do this. A lot of discussion that talks about a cut here or a boost there, but nothing concrete or widely accepted.
* Probably the biggest problem with using two tweeters is that reflections are rarely symmetrical. The reflected energy creates phantom imaging cues. It's like having a 2nd, 3rd, 4th and 5th set of tweeters. Because the cues are not symmetrical, the soundstage dimensions wander. This is particularly problematic when tweeters are mounted low, and the car is occupied by one person and not two. Absorbtion and diffusion of the reflected energy makes an audible improvement, as demonstrated by Eldridge in his 4Runner and by myself in the spherical enclosure thread.
Originally Posted by mitchyz250f View Post
Also, if you had the mids in the KP, you could get similar path lengths by placing the tweeters at the same PL towards the middle of the dash.
Am I stating the obvious or am I lost?
The perception of depth and width is due to interaural time delay, and we're not very sensitive to that above 1600hz. Due to that, I'm not certain that the tweeters need to be equidistant to the kick panels. If I were to use mono tweeters, I'd use time delay to sync 'em up with the midrange. Physically moving them is probably overkill.
I wonder how far you could take this. Perhaps a single supertweeter at the top of the windshield, time aligned to the midranges?
In summary, you ask if you used a single tweeter would the music sound like it was in mono. I'd reckon the answer is "no." Your perception of depth and width begins around 1600hz; as long as you locate the midranges and the midbasses carefully, you'll get your cues from them. This isn't to say that you should go home and turn off one of your tweeters however. Based on what I've read, tweeters *do* have a small influence on width and depth, but this is primarily due to the shape of the ear. And as soon as you put your tweeters too far off to the side, or even worse, the floor, you've ruined the illusion. Therefore, one tweeter in an ideal location may trump two tweeters in a less-than-ideal location.
I really need to build something like the JL car and find out.
You home audio people might be cringing
