Advanced Front Speaker Systems How-to

Excellent thread.  Here's another adddition for people designing such a system: installing a 3-way component set (or creating one as below) actually creates a net four-way system in your vehicle, with the following basic crossovers points:

1) The bottom three or so octaves are handled by the subwoofer, requiring a low-pass crossover at about 80 Hz.

2) The next 1.5 to 2 octaves (upper bass) are handled by the woofer (or mid-woofer) of the three-way.  This requires a bandpass crossover between the sub cross at 80Hz (high pass) and the upper cross at somewhere between 200 and 300 Hz (low pass.)  In some systems this crossover is another full octave higher on the order of 600-700Hz.

3) The next few octaves are handled by the mid-range.  This is another band-pass crossover between the higher point of the mid-woofer (200 to 800 Hz high pass) and the crossover to the tweeter, usually around 3500 Hz (low pass.)  Some tweeters play lower into the 1500 Hz range.

4)  The highest frequencies are handled by the tweeter and this speaker requires a high-pass crossover at the same frequency as the low-pass of the midrange.

In my car, I use a combination of electronic and passive crossovers.  I use a 24 db/octave electronic low pass for the subwoofer and an 18db/octave electronic bandpass for my midwoofer.  My midrange uses an 18 db/oct electronic high-pass, but then the crossover to the tweeter is passive.  These are the crossovers supplied with my 2-way component sets (18 db/oct @ 1700 Hz).  I created this 4-way system to suit my tastes, adding a mid-woofer between 80 Hz and 240 Hz in kick panels since I felt the woofer of my component set lacked detail in the lower region.

More ideas soon.  Thanks for starting this Steven!

Ok, now that I've covered speaker selection and how to build the front speaker system, I will now move on to installation. I'm going to break it down into two parts, speaker location and speaker aiming.

Speaker Location: It seems to me that there is a lot of misinformation out there on this subject. Usually people will say that kick pannel locations are best. The reasoning is that you want to reduce the pathlength differences between the two sides. To clarify that a bit; you want to have the distance between your head an the left speaker to be as close as possible to the distance between your head and the right speaker. Kick pannel installations do offer this advantage over the more common stock lower door mounting location. However the question is if this difference is audible. The answer is usually no.

To understand the reason why it isn't audible, you have to understand some things about how the human ear works. For instance, you must understand where we get our imaging cues from. For frequencies above 1000 Hz, we get our imaging cues strictly from differences in output. In other words, if a tone within this band of frequencies is played louder by the left speaker then the right, then the tone will sound like it is coming from the left side. For frequencies 500 Hz and down, imaging cues are strictly preceived by timing cues. Now you will probably notice that midrange speakers do play a portion of those frequencies effected by timing cues. So therefore those kick pannel locations should still make sense correct? Well there is one other phenomenon we must look at. That is the latent time of fusion. Basically what that is is how quickly our ears can preceive sound. Our ears cannot preceive sounds quicker than around 3 miliseconds. In other words if you played 2 tones within 3 miliseconds of eachother, they would sound like one tone. So to make an actual audible difference in the timing cues of a speaker, you would need to move it to where the sounds it produces are 3 miliseconds sooner or later than the original spot. This works out to be about 3 feet. So unless you are moving the speakers 3 feet from the stock locations, then you won't get an audible difference in timing cues.

So what can we learn from all of this? Well first of all, it is recommended to adjust the gains on the amp/processor to balance any differences in pathlengths. What this means is turn down the level of the closer set of speakers. Also if you have a phase/time alignment adjustability in any processor to utilize it to balance the timing differences as well. Once you get those set, and as long as you have a good frequency response, you should have an incredible soundstage with very focused imaging.

Speaker Aiming: Many people also recommend aiming the speakers toward the listener. In some instances, this could be benificial. However it is important to understand when to go throug the trouble to do it. Speakers produce sound quite evenly across a very wide axis. However as you start to go up in frequency, they will start to "beam" the sound. The frequency at which this starts to happen depends on the size of the speaker in question (D^2/lamda). A larger midrange like a 6.5 will start to beam at about 2.5 kHz. A 25mm (1") tweeter will start to beam at about 18 kHz. This doesn't mean that it doesn't produce any sound above those frequencies at a given off axis point. It just means that those higher frequencies are reduced. Many companies offer frequency response measurements of their speakers at different axis points. Take some time to look these over and see what the frequency response does when the speaker isn't pointed directly at you. Also note the frequency different sized drivers start to beam.

So when do we need to aim the speakers? Well if you are planning on using a 6.5" midrange up to 4 kHz, then it would be a very good idea to aim it more toward you. Otherwise you would have a dip in the frequency response that could have serious negative effects to equalize back. The more you aim it toward you, the less of a dip there will be. However if the driver isn't going to beam within the bandwidth you are running it, then there is no need to aim the driver. Also, if you are going with the passive crossover design mentioned earlier, then you can reduce this a bit at the cost off efficiency.

Hopefully that will help you guys install your speakers correctly. I do want to give credit where credit is due. Most of the stuff I have posted is what I have learned from my boss, Dan Wiggins. Again, please feel free to ask questions if you have any, add your own comments, or disagree. Keep in mind that there are no dumb questions.

Steven Kephart

Adire Audio

First of all, thanks guys for your comments.  I really appreciate it.

Now I want to expound on the above.  I was talking this over with my boss and learned some more interesting information that is important here.  From the above, the question comes up on whether or not you need time alignment processing with more precision than 3 miliseconds.  The answer is yes.  My comments above were specifically about how timing changes effect imaging cues.  However the timing changes do effect the frequency domain as well.  So although the imaging cues aren't effected (audibly), the frequency response is.  One thing I remember my boss mentioning as an example is that changing the timing of the speakers will effect the spectrum of the comb filtering.  This will have an impact on how the speakers sound.  So it is important to time align the speakers as precise as possible. 

And from there the question comes up about whether kick pannel locations would be benificial in the frequency domain.  Moving the speakers to the kick pannels will effect the frequency response, which will provide an audible difference.  However it is questionable whether it will be good or bad.  You are placing the speakers into deep chasms with quite a few more surfaces to create early reflections that will have a great effect on the sound.  The question you have to ask yourself with all this information provided is whether you are willing to give up your foot space, plus the money and time to build the kick pannels just for something that may be better, and could be worse.

Steven Kephart

Adire Audio

Would you recommend going Extremis/Tweeter, as the Extremis goes to 3K extremely well, and matching it with a tweet that can get down to 2.5 easily,

OR

Extremis/FR125?  If those FRs go to 22kHz the way they are saying...but would you have to cross the FRs over around 2-4kHz to avoid them both putting out the same band of frequencies?

With 3-way being better because you can focus more on a specific bandwidth per driver, that would mean the FR125 would be too close to the desired bandwidth of the Extremis, making a tweeter a better idea? Or no?

Also, would beaming become an issue running the FR125 too high, making a 2-way tweeter application a better idea?