DIYMobileAudio.com Car Stereo Forum banner

kickpanel speaker locations : pros, cons

146K views 231 replies 86 participants last post by  derickveliz 
G
#1 ·
I figured I'd start a thread about the pros & cons of kickpanel speaker locations, because there seems to be a lot of misinformation on the topic floating around this forum :( Please allow me to first say that I'm not a huge fan of kickpanel installations ... mostly because I find them to be ugly and intrusive ... and for the record I will absolutely admit that I have not heard the finest incarnation of the art, realized by some of the best installers/tuners in the biz. I hope to correct that sometime soon.

A brief review of dominant hearing mechanisms is in order. For the purposes of this discussion, it's VERY important to decompose hearing localization into two dimensions : lateral and vertical. I think the most important points are these :

1. Localization cues in the lateral plane are dictated by the differences in the acoustic transfer function to our two ears. These include inter-aural time differences, mostly significant for lower frequencies in the midbass and lower midrange, and inter-aural level differences, mostly significant for higher frequencies in the upper midrange and treble.

2. Localization in the vertical plane is dictated primarily by the shape of our outer ears and torso. Please consider that, if our outer ears and torsos were absent, we would be left with the classic "sphere with holes in the sides" for our heads ... and the symmetry of this geometry would absolutely dictate that vertical localization would be impossible. It's my understanding that the outer ear has the largest impact on the acoustic transfer function for vertical localization ... and given the dimensions of the outer ear, this implies that vertical localization cues are only significant above ~1kHz or so.

To summarize :

1. Ears are on the sides of our heads. Through ITD, IID we use both ears for left/right cues.

2. Vertical cues are given to us by the shape of our outer ears ... not vertically symmetrical, thank you, but common to both ears ... and the corresponding impact on the acoustic transfer function to our eardrums.

Conclusions : Let's say as an audio enthusiast, you happen to care about a realistic soundstage in your vehicle ... laterally as well as vertically. Furthermore, let's say you happen to care about a realistic soundstage for both front seat passengers, at the same time (For what it's worth, this ain't me ... I'm lazy, unskilled in fabrication, and selfish). Here's why kickpanels make sense in this case :

1. Left, right pathlength differences cannot be ELECTRONICALLY compensated for both front seat passengers at the same time. Simply not enough degrees of freedom, for the four ears involved. Kickpanels almost always present the best choice for PHYSICALLY equalizing PLD's.

2. Vertical cues can be electronically compensated, by a specialized form of equalization known as head-related transfer function inversion and substitution. A great reference on this topic is :

Creating Source Elevation Illusions by Spectral Manipulation, by P. Jeffrey Bloom, JAES, vol. 25, pp.820-828, September 1977.

So, simply put, the kickpanel optimization suggests that you solve the left/right problem physically, since it can't be solved electronically for both front seat passengers. The price you pay is in stage height ... but this can be solved electronically for both front seat passngers :) And yes, the best installers and tuners have known all this for many, many years.

Incidentally, there's another interesting conclusion to be drawn : physical separation of drivers in any single channel ... like mid & tweet, for example ... is not necessarily the horror of horrors it's often considered to be, particularly if the separation is vertical. Either the ear can't tell to begin with, since vertical localization cues don't even start until the treble, or spectral manipulation (EQ) can be employed to very convincingly trick the ear. Check the AES reference above. One of the most convincing soundstages I've ever heard ... with no obvious penalties in tonality, or coherence ... had midranges in the kicks and tweets in the A-Pillars. And I'm far, far from alone on this one ;)

Armed with that background, perhaps we can list some kickpanel pros & cons :

Pros :

- Most convincing lateral soundstage for both front seat passengers at the same time.

- Early reflections ... floor, underdash ... are relatively easy to tame, compared to windshield reflections of higher-mounted drivers, for example.

Cons :

- Often ugly and intrusive (to me, anyway)

- Establishing vertical stage height requires clever EQ ... spectral manipulation, involving HRTF inversion and substitution.

- Early diffraction caused by feet and legs.

Now it may be, that some users don't care at all about staging ... only tonality. There's plenty of room for everyone in this hobby :)
 
See less See more
#52 ·
bdubs767 said:
thats my debate right now, I can't make up my mind wheter to re do my kicks with the hiqophon 0w1-fs and try soem new angles for more height or jsut add the second tweet...

argghhhh

is there anyway you could put one there to test it out before the full install?
 
#56 ·
not at all. the AS2K stuff is pretty well written and easy to understand. i am not an engineer at all, but see audio very much as a hobby and as a way to get more insight into the world i live in. the tech briefs are invaluable and are a great read for even a complete novice. all you need is a tiny bit of patience and an open mind and they will help a lot in getting some grounding on what makes this stuff work. i still have all of mine from my original subscription more than 10 years later. good stuff! too bad they stopped publishing.
 
#57 ·
yeh, i am switching to 4.5" revelators since my 7" ES-07's are way to big for kicks lol
 
#58 ·
vactor said:
i am not an engineer at all, but see audio very much as a hobby and as a way to get more insight into the world i live in.
funny that you say that cause i am going to school to be an engineer. though with school, my job, and trying to finish my never ending install i dont think i would have time to read it :(
 
#59 ·
i know it's been a loooong time but i just did something like this last weekend and then I saw this thread. i have a set of Dyn 340s with the mid and tweeter in the kicks and I've NEVER been happy with the top end. it had no sparkle unless i turned the tweets WAY up and then it was just poop. well i had a tweet go out and i'm waiting for the replacement. i've has a set of .75" CDT metal domes (more on those later) in the OEM sail panel positions in my Passat that i was going to use as accents to bring the stage up already in place but not wired so i disconnected the other good MD100 and wired the CDTs to the Dyn XO. WOW! I've always thought I didn't like metal domes but these kick the snot out of the MD100s in the kicks. other than them being to bright which I tamed a bit on the XO and turning the mid up (which was a mistake) and then using a shelving filter on the PPI DCX-730 i'm very happy with them. seriously, i'm almost tempted not to reconnect the MD100s. if i do reconnect them i know this, it won't be in the kicks. hell, i'm even tempted to lose the MD140/2s as well and just do a 2-way with the MW160s and one of the tweeter sets.
 
#60 ·
I fully agree on the original post, I have kicks in my Dakota, the arrangement in the civic with the dead pedal and me wanting to keep a stockish look pushed me to doors and sail panels. I like both, yea, they sound different. I LOVE the stage height of having the tweets higher and I DO use TA and it makes a TON of difference on and off. Regrdless of what EQing I used in the truck I could never fake my ears inbto the right stage height, even temporarily using 10 BANDS PER CHANNEL of Orban FULLY PARAMETRIC EQ, that's over $5K of broadcast EQ folks! It still sounds great but the height is not there. My Caddy had kicks and it sounded great, stage height was fine, a little low but fine, and the localization was AMAZING, BIG car though! I'm happy with the Civic, Kicks would hinder my driving expierience WAY too much. But as the OP said, there IS SOOOOOO MUCH to equalizing pathlengths that one should NEVER overlook kick panels as a option.

Chad (Bet that one has a bunch of typos!)
 
#61 ·
I had my time with dyn dome mids and finally gave up on them after getting things right, it took a second set of tweeters in the A pillars to get the height and center image right. I thought the tweets were great on their own but now I know there are far better choices. I am to swamped to go into great detail here but those where the last dome mids I have played with and have no desire to try any others, ever;)

It is so easy to build great imaging kick panels and if to much effort then just going with pillar tweets and door midbasses can do wonders if done right, very very easy to do!

I will never run passives in a car again, I have gotten lucky and had good results but so easy to tune an active setup and cost effective if done right(DIY drivers all the way!)

Rick
 
#62 ·
chad said:
10 BANDS PER CHANNEL of Orban FULLY PARAMETRIC EQ
Was it a baby blue 2 rackspace model...#622 maybe? If so, that's a hoot that you ran that in your car! We had one in school, and I remember getting some good electric guitar sounds from it.
 
#63 ·
First of all, let me say that I am very appreciative of the intelligence that you bring to this sport Werewolf. I am also appreciative of how you share that intelligence with the rest of us.

My question, for my own shelfish purposes is, do you prefer kicks or not? I have always had tweet mounted kicks in 2-ways and tweet and mid mounted kicks in three ways. I am currently looking to build kicks for my mids and tweets. I have read this post a couple of times and though many, less knowledgeable than you have strong opinions based on limited information, you don't seem to express a preference that I can see? Care to share?
 
G
#64 ·
see that ... it was my (self-assumed) responsibility to present the pros & cons without expressing my personal preference :)

But here it is. First, the qualification that i have NOT heard the best kickpanel incarnations out there. But I have heard many, and my simple conclusion is this : mid-bass drivers in doors, mids in kicks and tweets up high represent the best compromise for my personal taste :) Tweets down low cause a low stage in my experience, that the best tuning i've heard can't completely eliminate. My attention is simply drawn downward by low-mounted tweeters.

Midrange drivers in kicks makes sense ... logically and sonically, based on how we hear and what localization cues exist in the midrange. All the ITD info discussed in this thread regarding L/R and phantom center cues, for both front seat passengers, is best realized by maximizing pathlengths while minimizing pathlength differences. And when you realize the height insensitivity of our hearing in the midrange, kicks is where the midrange belongs. And in my limited experience, you just can't get stage depth unless you get those drivers in front of you ... as far in front of you as the car allows ... and put them where midrange reflections are best tamed. In other words, not only are L/R ITD cues sub-optimal with mids up high, but the nasty glass reflections also conspire to really limit stage depth.

Physical separation of midrange drivers and tweeters is a great concern to many ... but i think it's mispalced. First, it simply ignores the huge difference between lateral and vertical separation. Second, it ignores vertical insensitivity until treble frequencies (your ear has a real tough time telling where a midrange driver is in the vertical plane, simply because there's no hearing mechanism to locate it vertically ... and if your ears can't tell, who cares what your eyes think). Finally, it does not recognize the issue of understanding the acoustic signal that arrives at your eardrums as being more fundamentally important than how or where it originates. There is simply no direct comparison to home audio, because (as we've explored), there has been no compelling reason to explore ... and exploit ... these relationships in home audio. In short ... why does it matter if the mid and tweet are separated by many xover wavelengths, if the the signal that arrives at your eardrums (and all associated localization cues, horizontally and vertically) can't tell?

So there you have it, in a nutshell. What i've heard agrees with what acoustic principles tell me ... it's all a compromise (taming reflections with tweets up high, for example), but mids in kicks (paying most attention to L/R ITD and stage depth) and tweets up high (where the vertical cues are) is the best compromise I've heard :)
 
#65 ·
Mr Perfect said:
Was it a baby blue 2 rackspace model...#622 maybe? If so, that's a hoot that you ran that in your car! We had one in school, and I remember getting some good electric guitar sounds from it.
642B (has the balanced I/O module)

Top blue one...


I have another sitting on a shelf ;)

Chad
 
#66 ·
werewolf said:
Midrange drivers in kicks makes sense ... logically and sonically, based on how we hear and what localization cues exist in the midrange. All the ITD info discussed in this thread regarding L/R and phantom center cues, for both front seat passengers, is best realized by maximizing pathlengths while minimizing pathlength differences.
Can you explain the first of those two goals? Minimizing pathlength differences is fairly straightforward, but why is maximizing pathlengths important?
 
G
#67 ·
Here's how I look at it Mark ...

Let's first assume that we agree with the premise that we care about both front seat passengers at the same time (that's far from universal, i might add, but it is important to some).

By "pathlength differences", we typically mean the pathlengths measured from a left speaker to a listener, compared to the right speaker to that same listener. Let's say that all we cared about was minimizing pathlength differences ... that's our only criterion for optimization. Well, that might lead to putting a left speaker on the far left corner of the dash, and a right speaker on the center of the dash. Know what I mean? Pathlength differences are minimized ... in fact, PLD's can be made zero :) Optimization satisfied.

But alas, we would have two problems (at least) with such an optimization that seeks to only minimize PLD's. First, the stage will not be as "expansive" as it could be, for even our single listener. Second, the stage will be horrible for our other front seat listener, because minimizing PLD's for a single listener really created a sub-optimal situation for the other listener up front.

So ... minimizing PLD's itself does not, in my opinion, capture all of what we are trying to do. Only in conjunction with the other criterion ... namely, maximizing pathlengths ... do we maximize stage expansion, and tend towards an optimum configuration for both front seat passengers.

Make sense?
 
#68 ·
werewolf said:
Here's how I look at it Mark ...

Let's first assume that we agree with the premise that we care about both front seat passengers at the same time (that's far from universal, i might add, but it is important to some).

By "pathlength differences", we typically mean the pathlengths measured from a left speaker to a listener, compared to the right speaker to that same listener. Let's say that all we cared about was minimizing pathlength differences ... that's our only criterion for optimization. Well, that might lead to putting a left speaker on the far left corner of the dash, and a right speaker on the center of the dash. Know what I mean? Pathlength differences are minimized ... in fact, PLD's can be made zero :) Optimization satisfied.

But alas, we would have two problems (at least) with such an optimization that seeks to only minimize PLD's. First, the stage will not be as "expansive" as it could be, for even our single listener. Second, the stage will be horrible for our other front seat listener, because minimizing PLD's for a single listener really created a sub-optimal situation for the other listener up front.

So ... minimizing PLD's itself does not, in my opinion, capture all of what we are trying to do. Only in conjunction with the other criterion ... namely, maximizing pathlengths ... do we maximize stage expansion, and tend towards an optimum configuration for both front seat passengers.

Make sense?
If I may paraphrase a bit of that for us ADD type readers :D -the farther forward you can put the drivers the better, correct?
 
#70 ·
werewolf said:
Here's how I look at it Mark ...

Let's first assume that we agree with the premise that we care about both front seat passengers at the same time (that's far from universal, i might add, but it is important to some).

By "pathlength differences", we typically mean the pathlengths measured from a left speaker to a listener, compared to the right speaker to that same listener. Let's say that all we cared about was minimizing pathlength differences ... that's our only criterion for optimization. Well, that might lead to putting a left speaker on the far left corner of the dash, and a right speaker on the center of the dash. Know what I mean? Pathlength differences are minimized ... in fact, PLD's can be made zero :) Optimization satisfied.

But alas, we would have two problems (at least) with such an optimization that seeks to only minimize PLD's. First, the stage will not be as "expansive" as it could be, for even our single listener. Second, the stage will be horrible for our other front seat listener, because minimizing PLD's for a single listener really created a sub-optimal situation for the other listener up front.

So ... minimizing PLD's itself does not, in my opinion, capture all of what we are trying to do. Only in conjunction with the other criterion ... namely, maximizing pathlengths ... do we maximize stage expansion, and tend towards an optimum configuration for both front seat passengers.

Make sense?
Sure. But let's assume for a moment that we don't give a rat's ass about the passenger. ;) When that's the case, is it beneficial to maximize pathlengths? Also, would TA sufficiently take care of PLDs if we don't care about the passengers? I can't think of why it wouldn't, except perhaps if the pathlength difference also carried with it a big difference in reflections and dispersion between the two speakers.
 
#71 ·
I believe that longer pathlengths in a car may create a situation where the intensity of direct sound vs. reflected sound is less severe. That tends to give the impression of more space.

I don't think pathlength differences are all that important, especially with time alignment so readily available. However, having equal pathlengths does help mitigate alot of problems besides time arrival, such as intensity, and ratio of direct to reflected sound at the listening position provided the drivers are aimed similarly toward one position. Which is one of the reasons why I don't like kickpanel installs where the drivers are aimed in a crossfiring manner in an attempt to create a stage for both passengers... in many cases you will have the passenger side speaker aimed nearly at the driver, but the driver side speaker about 45 degrees off-axis from the driver which I don't think creates an optimal listening environment for the driver.
 
#72 ·
I think thats what also drives the use of smaller speakers for kickpanel locations.

The crossover points we tend to use combined with smaller drivers let us use aiming positions that can get pretty radical without affecting the *anechoic* frequency response on each side!

much less when both sides are slightly toed in, minimizing the angle/aiming differences per side.
 
G
#73 ·
I agree with everything you guys have just said, with the one caveat being ... if you care about both front seat passengers at the same time, you'll have a sub-optimal solution (stage-wise) if you don't maximize pathlengths and minimize PLD's.

In the extreme (engineers tend to think in limits, ya know) ... minimizng PLD's for a single passenger, without regard to maximizing pathlengths, leads to ... headphones! I know, not exactly ... but you guys know what I mean.

The point about direct vs. reflected energy is a very important one too, i think. And i strongly suspect that's why you will never realize the full potential of stage depth in a car without maximizing pathlengths.
 
#74 ·
I don't understand why the reflected sound would be more exaggerated in the short-pathlength situation. In fact, I would have thought the opposite to be true. If the speaker's closer, it's louder. The reflecting surfaces, on the other hand, may or may not be closer or in general more apt to cause reflections. An example using limits (although not quite as severe as werewolf's example) would be putting your ear right up to the speaker. You'd expect reflections to play less of a role in that case.

Again, I'm not seeing how pathlengths make any difference to a time-aligned level-adjusted system. Reflections should certainly be affected, but it's not clear to me that it's for the worst if the speaker is brought closer to the listener.
 
#75 ·
Whiterabbit said:
I think thats what also drives the use of smaller speakers for kickpanel locations.

The crossover points we tend to use combined with smaller drivers let us use aiming positions that can get pretty radical without affecting the *anechoic* frequency response on each side!

much less when both sides are slightly toed in, minimizing the angle/aiming differences per side.
In other words, a Bose 901-like ratio of direct to bounced sound. Sounds bloody wonderful to me...
 
G
#76 ·
admittedly, i've confused a couple points ...

As we've explored, one of the downsides of mounting drivers "up high" is that, in a car, "up high" means a more reflective local environment ... more glass up high, no way around it. Whereas kickpanel mounting allows for the possibility of some (perhaps limited, given midrange wavelengths, but not altogether uneffective) treatments of local reflecting surfaces ... under dash, floor, kicks, etc.

So I would say that mounting drivers in the kicks has a better chance of controlling short-time, local reflections. In this regard, kicks minimize PLD's, maximize pathlengths and allow for some amount of reflection control ... all leading to optimized stage width and depth cues. To be sure, the reflection control is more effective for tweets in the kicks, but if you put the tweets up high, while reflections are still something to certainly deal with in the treble, at least you've got the drivers important to vertical localization where you need them.

If anyone has additional info on maximizing pathlengths to help minimize reflected energy (or maximize the ratio of direct-to-reflected), please feel free to post away :)

Yes ... a compromise to be sure. But all factors considered, it's pretty much my favorite :) At least for conventional drivers ... horn-loaded compression drivers is a whole 'nother story, perhaps for another thread. Somebody else can take the ball on that one ... :p
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top