Dayton+Seas door install - Caddy STS

[MarkZ]

Got to my front doors this weekend, and finished them in the bitter cold today. I installed a Dayton RS180 7" and Seas P11RCY 4.5", replacing an a/d/s/ 3-series 6.5" and 4". The goal is to keep a stealth installation and maintain the stock look.

So here's the old installation:

http://mdz.no-ip.org/audio/install/speakers/old/newfar-sm.jpg

I kept the a/d/s/ 3-series tweeter, which is currently mounted in the stock locations in the dash (not pictured). I really like the smooth yet vivid sound of this tweeter, so there are no plans to replace it in the near future.

So I quickly came to the realization that under no circumstances would I be able to fit both the Dayton and Seas behind the grill as I did with the a/d/s/ components. The only other option was to mount the Seas somewhere higher in the door.

First, I installed the Dayton on a 1/2" MDF baffle in the lower door:

http://mdz.no-ip.org/audio/install/speakers/daytondoor-sm.jpg

Given window-clearance and door panel concerns, there was only one place to put the Seas driver and that was behind a padded panel above the door handle. I removed the entire door panel and mounted the driver to a dynamat-covered abs plastic baffle attached to it, and sealed the back of the speaker from the front by use of a large PVC cap (mounting depth wasn't much of a concern in this location). Here you can see the back of the outer door panel and the PVC cap:

http://mdz.no-ip.org/audio/install/speakers/pvc-sm.jpg

Here's the door with the speakers and main door panel back in place:

http://mdz.no-ip.org/audio/install/speakers/dooropen-sm.jpg

Then I removed a 4.5" circle of foam and plastic backing on the padded panel -- the thin perforated covering remained:

http://mdz.no-ip.org/audio/install/speakers/panelback.jpg

http://mdz.no-ip.org/audio/install/speakers/panelfront-sm.jpg

After reinstalling the padded panel and stock grill, here's the finished product:

http://mdz.no-ip.org/audio/install/speakers/door-sm.jpg

My original concerns were twofold. Is the mid mounted too far back in the door and will this have implications for the soundstage? I've done upper door installations in the past with good results, but the speakers were a little closer to the front of the door.

My second concern was with just how acoustically transparent that perforated leather stuff was going to be. Ultimately, I decided not to worry about it much. I tested it in the house by putting a tweeter behind it and one in front of it and couldn't distinguish an obvious difference. Also, any possible high freq attenuation that it caused would still be less than the attenuation that accompanies the driver-side door installation.

[alphakenny1]

looks good man! my friend had the 2 series a/d/s. its pretty decent but probably not as good as your new dayton, seas combo! good job.

[npdang]

Awesome install. I like the idea of putting the midrange up high, behind the perforated leather. That's the first time I've seen that :)

[newtitan]

wow thats very creative , mids under leather in a CADDY to die for

love the CTS

whats the trunk look like?

[MarkZ]

wow thats very creative , mids under leather in a CADDY to die for

love the CTS

whats the trunk look like?

I'll have pictures of the trunk hopefully by Monday. I have to play around with wires a bit this weekend and remount some of the power accessories. But the trunk essentially consists of a 9-space rack with three Behringer pieces, a rackmount computer, and the audio equipment. The rack and sub box are behind a false wall.

[jisturm]

Wow. Nice. I love the install and would love pics of the rest of the ride!

[MarkZ]

*Update*

Changed the installation a bit. Re-sealed doors, created new baffle for mids, added Morel MR55 dome midrange (similar to this (http://www.renaudio.com/carstereo/MR-56.htm) one) to handle ~1.5kHz-5kHz.

[MarkZ]

Doors are starting to get a little chewed up, but here's the latest.

Peerless Exclusive mid, Morel dome, Adire Extremis 6.4. a/d/s/ tweeter remains in the dash, but I'm thinking of trying a Vifa XT25 in its place.

[rcurley55]

Am I the only one who feels like this is going in the total wrong direction? I admire your creativity, but from a sonic perspective, you are going backward quicly imo...

[MarkZ]

Am I the only one who feels like this is going in the total wrong direction? I admire your creativity, but from a sonic perspective, you are going backward quicly imo...

Care to elaborate?

[chuyler1]

The speakers are too close to you, further skewing the left/right bias. On top of that, leather isn't exactly acoustically transparent.

I admire your creativity but there are reasons people install speakers down low and as far away from the listener as possible. Even with time alignment, your near sided speaker is going to be much louder than the far sided speaker and your sound stage is going to be in your face (instead of out in front of you).

[MarkZ]

The speakers are too close to you, further skewing the left/right bias.

Agreed. That's the biggest drawback of that installation. However, attenuating the left speakers and introducing time alignment has helped tremendously.

On top of that, leather isn't exactly acoustically transparent.

One would have thought. That's why I RTA'd them with the door skin installed and not installed. Initially I did this to try to figure out a good lowpass cutoff point, since I figured it would rolloff the speakers at high frequencies. Interestingly enough, it had very little effect when the door skin was in place at the cutoffs I was using anyway (~3kHz). I'll soon have measurements to post.

I admire your creativity but there are reasons people install speakers down low and as far away from the listener as possible. Even with time alignment, your near sided speaker is going to be much louder than the far sided speaker and your sound stage is going to be in your face (instead of out in front of you).

"Down low" wasn't an option. I refused kickpanels in this car from the beginning, as the primary goal was stealth and lack of intrusion on my feet (and therefore my comfort). I've gone the kickpanel route many times before, and I felt it inappropriate in this car. So my first installation actually went in the lower doors. I tried to fix the inherent problems of leg occlusion by playing with angling, but to no avail. My only option then was to mount the speakers higher, and this was the only way I could pull it off without COMPLETELY redesigning the entire doors -- no easy (or cheap) feat, especially when your upholstery is leather.

[lpreston]

With the drivers above the door handle, does this not shift your stage rearwards?

[MarkZ]

With the drivers above the door handle, does this not shift your stage rearwards?

Even though they're above the handle, they're still well forward of where I sit. But, as chuyler mentioned, the configuration doesn't put the soundstage far out in front of me, which of course would be ideal. But it also doesn't pull it as far to the sides as you might expect from the pictures. A byproduct of trying to keep the install stealth and non-intrusive, and to even out the FR by putting them higher up. Some of you already know how I feel about the relative importance between soundstage and FR and which one to choose over the other when necessary. ;)

[rcurley55]

Care to elaborate?

It's already been touched on before, but here you go:

1. you have gone backward in the case of speaker location - you have completely destroyed the concept of minimizing pathlength differences between your left and right side in the ever-so-critical midrange frequencies. I personally don't think any amount of ta or level matching is going to get you the spacial presentation that you could have gotten from just a simple two way with midbasses in the doors and tweets in the sails or pillars.

2. you threw the idea of KISS out the door - you have added more speakers, which generally, due to phasing, multiple point sourcing, etc is a bad idea. Why you need two midrange speakers (assuming the dome and the other installed behind the arm rest are both being bandpassed) is a very bad thing imo.

3. despite your rta measurements, I'm highly suspect of leather (perforated or not) being acoustically transparent.

Like I said, while I applaud your creativity, your execution (for lack of a better term) sucks. Sorry, but I'm not going to sugar coat it ;)

I just think you could have done much better with a very simple, fundamentally sound approach (for less money and equipment mind you) than your current approach. Then again, it's not my car, so at the end of the day all that matters is your happiness.

[MarkZ]

Thanks for the comments. Response/rationale below.

1. you have gone backward in the case of speaker location - you have completely destroyed the concept of minimizing pathlength differences between your left and right side in the ever-so-critical midrange frequencies. I personally don't think any amount of ta or level matching is going to get you the spacial presentation that you could have gotten from just a simple two way with midbasses in the doors and tweets in the sails or pillars.

Three comments to the above. First, I don't agree with you that processing can't make up for the minimization of pathlength differences. I'd like to hear why you don't think it can.

Second, a two-way introduces a whole host of problems that are FAR worse than any having to do with pathlength differences. Among these problems:

* ankle-pointing midrange is exactly what I was trying to combat with the upper door installation. Been there, done that (as you can see with the first picture in the thread). IMO, it's more difficult to compensate for leg occlusion and the huge difference in FR that arises with the different aiming between left and right, than it is to compensate for pathlength differences.

* a two-way is INCAPABLE of midbass reproduction at appreciable output levels. I know this sounds extreme, but I've never heard midbass pulled off correctly in a two-way system at the output levels I'd like the system capable of achieving -- at least not with cone sizes that can reasonably be fit in a door. This is to be expected, because tradeoffs must be made in one realm (excursion, sensitivity, and high frequency response) in order to improve the other. Put simply, when you improve low frequency sensitivity and output capabilities, you tend to sacrifice high frequency performance. If I were to simplify to a two-way, I'd likely go the route DS-21 has gone and use an SLS 8" with a widebander like the Aura that he uses. But IMO that'd be a downgrade too, because the SLS doesn't quite fit my needs (although it's close) and I'd be sacrificing high frequency response -- PLUS I'd still be left with the pathlength issue!

* a two-way would require a tweeter that could go low, but my dash locations preclude the possibility of using large format tweets, which make that a difficult prospect.

And third, in keeping with the primary installation goal of "stealth and stock", tweeters in the pillars wouldn't work. If I could choose to upgrade anything, that would be exactly what I'd do. But I've gone to great lengths to avoid modification in places that you can see, and going into the installation this was a primary goal. Even though I don't hate the tweeters being in the stock locations (windshield firing), I agree with you that pillars would be better.

2. you threw the idea of KISS out the door - you have added more speakers, which generally, due to phasing, multiple point sourcing, etc is a bad idea. Why you need two midrange speakers (assuming the dome and the other installed behind the arm rest are both being bandpassed) is a very bad thing imo.

This is the most common objection to my installation, and rightfully so. First of all, it's important to note that there isn't really a multiple point sourcing problem with the additional midrange since it's literally right next to the other one -- unless you're referring to potential lobing at the crossover frequency, but I don't think that's an issue at the frequencies we're talking about. Anyway, the rationale behind the dome mid originally is because the Seas mid I was using couldn't extend high enough to meet the tweeter. Its response above 1kHz or so was pretty poor, IMO. Yet the dome mid couldn't be used alone because I was using purely a bass driver as the midbass (Peerless XLS first, and now Adire Extremis). So I introduced the dome (which I had lying around, actually) which cleared up my problem and cost me nothing.

I've tried it both with and without the dome mid, which I can do with the touch of one button to recall a preset since I use a pair of DCX2496's for my processing. I determined that using the dome mid was a clear advantage. However, I haven't yet performed the same test with the Peerless Exclusives I just installed. I have the feeling that they can go higher than the Seas driver. High enough to meet the 3.5kHz xover point? We'll see.

3. despite your rta measurements, I'm highly suspect of leather (perforated or not) being acoustically transparent.

I didn't mean to make it seem like it was. My point was that the shape of the RTA curve did not change with it in place. There seemed to be some broadband attenuation, but it didn't introduce a transfer function that I couldn't overcome with simple gain adjustments.

[rcurley55]

Mark - as I said before - if you are happy, that's what counts. I'll just make a few comments.

First, I don't agree with you that processing can't make up for the minimization of pathlength differences. I'd like to hear why you don't think it can.

I shouldn't have said, just pathlength differences - because, I could put my left and right speakers next to eachother, and boom, zero path length differences. What you have done is caused your path lengths to be VERY short. Then electrically compensated one to still be short. In essence, as you have admitted, you have little or no depth to stage.

Hence, a lack of spatial presentation.

Second, a two-way introduces a whole host of problems that are FAR worse than any having to do with pathlength differences. Among these problems

I guess that depends on how much stock you place in the technical representation of your music.

* ankle-pointing midrange is exactly what I was trying to combat with the upper door installation. Been there, done that (as you can see with the first picture in the thread). IMO, it's more difficult to compensate for leg occlusion and the huge difference in FR that arises with the different aiming between left and right, than it is to compensate for pathlength differences.

I've had door mounted mids (two way system) and have not had major issues while driving - I'm 5'9", so perhaps these issues affect you differently.

Different aiming between left and right is a necessity often in car audio - especially the off axis nature of many people's installs - typically this is how car audio drivers are meant to be used, and it actually helps to compensate for path lenght issues....

* a two-way is INCAPABLE of midbass reproduction at appreciable output levels. I know this sounds extreme, but I've never heard midbass pulled off correctly in a two-way system at the output levels I'd like the system capable of achieving -- at least not with cone sizes that can reasonably be fit in a door.

I bolded a potential issue...

* a two-way would require a tweeter that could go low, but my dash locations preclude the possibility of using large format tweets, which make that a difficult prospect.

you don't need a large format to get low...

And third, in keeping with the primary installation goal of "stealth and stock", tweeters in the pillars wouldn't work. If I could choose to upgrade anything, that would be exactly what I'd do. But I've gone to great lengths to avoid modification in places that you can see, and going into the installation this was a primary goal. Even though I don't hate the tweeters being in the stock locations (windshield firing), I agree with you that pillars would be better.

I've seen lots of steathy installs in pillars - are you sure it's not possible? Not questioning your ability - just not sure if you've looked....give it a try :)

Anyway, the rationale behind the dome mid originally is because the Seas mid I was using couldn't extend high enough to meet the tweeter. Its response above 1kHz or so was pretty poor, IMO. Yet the dome mid couldn't be used alone because I was using purely a bass driver as the midbass (Peerless XLS first, and now Adire Extremis). So I introduced the dome (which I had lying around, actually) which cleared up my problem and cost me nothing.

Just "cost" you extra:
-runs of speaker and rca wire
-extra amplifier channels
-extra processor power
-extra current on your charging system
-extra tuning issues

so while it may not have cost you anything monetarily - certainly it "cost" you something :D

If it makes you happy - that's great - your approach is not what I would call technically sound for a good audio system. You are making things too difficult in my opinion - but then again, we are on a public message board, so take my advice for what you paid for it - nada ;)

[MarkZ]

I shouldn't have said, just pathlength differences - because, I could put my left and right speakers next to eachother, and boom, zero path length differences. What you have done is caused your path lengths to be VERY short. Then electrically compensated one to still be short. In essence, as you have admitted, you have little or no depth to stage.

Hence, a lack of spatial presentation.

Fair enough. Car audio is about tradeoffs. And at the risk of sounding like a broken record, it's my opinion that spatial presentation is too highly emphasized in the field, to the point where a lot of people suggest sacrificing things like FR and output capabilities (and hence, introducing distortion) to achieve it. I'm not willing to make those sacrifices, especially since a good portion of the music I listen to makes very little effort to record in a manner that would take full advantage of spatial listening. In other words, I think imaging is overrated. :D

I've had door mounted mids (two way system) and have not had major issues while driving - I'm 5'9", so perhaps these issues affect you differently.

That's no easy feat, then. When you're listening to one speaker 60 degrees off axis and your legs are occluding it (which is absolutely the case for me in my vehicle), and the other speaker is on-axis and unoccluded, you're gonna have some pretty nasty FR effects going on. These are very difficult to compensate for, and you still typically have L-R pathlength issues.

I bolded a potential issue...

More importantly, I explained why it's IMPOSSIBLE to have a speaker that does it all. Do you know of a speaker with under ~3.5" mounting depth and same or smaller footprint than the XLS that will give me the output capabilities of an Extremis or XLS down to 70 or 80 Hz, and at the same time extend up to 3kHz or so to meet the tweeter?

you don't need a large format to get low...

You won't find very many small tweeters that can do it. But I suppose they're out there.

I've seen lots of steathy installs in pillars - are you sure it's not possible? Not questioning your ability - just not sure if you've looked....give it a try :)

Perhaps "stealth" is not the best word I could have used. How about "visible modification"?

Just "cost" you extra:
-runs of speaker and rca wire
-extra amplifier channels
-extra processor power
-extra current on your charging system
-extra tuning issues

so while it may not have cost you anything monetarily - certainly it "cost" you something :D

Nah, the addition of the Morel didn't cost me anything. I had the amplifier channels, wiring (got tons of it), processors, and the extra current is negligible.

If it makes you happy - that's great - your approach is not what I would call technically sound for a good audio system. You are making things too difficult in my opinion - but then again, we are on a public message board, so take my advice for what you paid for it - nada ;)

I appreciate the input. It's sparked discussion, which is what this forum is all about.

You bring up an interesting point about "technically sound". I don't necessarily agree with your assessment. In home audio, for example, 4-way systems are abundant. However, in home audio, people typically don't place the subwoofer in another room or have issues related to cabin gain and the like -- basically, in home audio, you can have a single low frequency driver take care of your sub bass as well as a good portion of your midbass. If I could fit subwoofers in my doors without modification, I would, and then I wouldn't need dedicated midbass drivers. The additional low frequency drivers in the doors are necessary for my application, and unfortunately they don't extend as high as I'd like. With the parts available to me at the time, I was able to take care of those high frequencies with the Seas/Morel/tweeter combination. I just put the Exclusives in today, so like I said in my last post, they may eliminate the need for the Morel domes.

Regardless, based on the confines of my installation, kick panels and dash pods are out the window, and moving the window motor was not an option, which left that location in the upper door the only possibility for midrange.

[OldOneEye]

Thanks for the comments. Response/rationale below.


I didn't mean to make it seem like it was. My point was that the shape of the RTA curve did not change with it in place. There seemed to be some broadband attenuation, but it didn't introduce a transfer function that I couldn't overcome with simple gain adjustments.

I just wanted to add, would an RTA catch any anomalies you might hear from the sound bouncing back off the leather towards the speaker? Also, while the shape of the curve didn't change, how about the level (so to get the same amount of sound, you will have to drive the speaker harder, making it more likely to have less headroom and more likely to have the speaker be closer to the point where it is not increasing its output in a linear fashion)?

Just curious, what kind of pathlenth do you have now compared to before.

Juan

[MarkZ]

I just wanted to add, would an RTA catch any anomalies you might hear from the sound bouncing back off the leather towards the speaker? Also, while the shape of the curve didn't change, how about the level (so to get the same amount of sound, you will have to drive the speaker harder, making it more likely to have less headroom and more likely to have the speaker be closer to the point where it is not increasing its output in a linear fashion)?

Yeah, that's one mechanism I'd expect (bouncing off thing), even though it's probably not the most reflective surface in the world. But one would think that it would create distortion rather than major FR issues. Would you be able to pick it up on an RTA? Only if the distortion was substantial.

Anyway, the amount of attenuation it caused was surprisingly low. Like I said in my first post in the thread several months ago after I first tried this strategy, I was expecting it to be a real problem (which is why I initially rejected the idea altogether). But actually implemented, it turned it that it didn't attenuate things a whole lot. I don't have an exact number to give you, although this entire discussion is persuading me to go out and do a more rigorous measurement to post to you all.

FWIW, replacing those Seas drivers with the Peerless ones actually resulted in more attenuation than the door covering. The difference in efficiency is sizable.


Just curious, what kind of pathlenth do you have now compared to before.

The speakers aren't really as close as some of you might think. I was actually checking things out this morning (again, prompted by the suggestions and comments - thanks, guys!), and the difference in path length to my head from those midrange locations is about the same as a tweeter at eye level in the pillars would be, if that gives you any idea. They're still kinda low and forward (about 45 degrees off-axis). But fortunately no part of my body is in the way, and ditto for when there are passengers unless they're leaning far forward.

[bassfromspace]

Yeah, that's one mechanism I'd expect (bouncing off thing), even though it's probably not the most reflective surface in the world. But one would think that it would create distortion rather than major FR issues. Would you be able to pick it up on an RTA? Only if the distortion was substantial.

Anyway, the amount of attenuation it caused was surprisingly low. Like I said in my first post in the thread several months ago after I first tried this strategy, I was expecting it to be a real problem (which is why I initially rejected the idea altogether). But actually implemented, it turned it that it didn't attenuate things a whole lot. I don't have an exact number to give you, although this entire discussion is persuading me to go out and do a more rigorous measurement to post to you all.

FWIW, replacing those Seas drivers with the Peerless ones actually resulted in more attenuation than the door covering. The difference in efficiency is sizable.




The speakers aren't really as close as some of you might think. I was actually checking things out this morning (again, prompted by the suggestions and comments - thanks, guys!), and the difference in path length to my head from those midrange locations is about the same as a tweeter at eye level in the pillars would be, if that gives you any idea. They're still kinda low and forward (about 45 degrees off-axis). But fortunately no part of my body is in the way, and ditto for when there are passengers unless they're leaning far forward.

So you're saying that in essence, you can't ride around with any hunchbacked ladies in your passenger seats? :D

[MarkZ]

So you're saying that in essence, you can't ride around with any hunchbacked ladies in your passenger seats? :D

They ride in the trunk.

[honfatboy]

Cheers to MarkZ for taking the feedback and responding throughly and objectively, even when rudely, and somewhat unnecessarily, criticized. In my experience, it's not often that someone will respond calmly and maturely to an opposing view.

[khail19]

Cheers to MarkZ for taking the feedback and responding throughly and objectively, even when rudely, and somewhat unnecessarily, criticized. In my experience, it's not often that someone will respond calmly and maturely to an opposing view.

Agreed, but I don't think anyone was rude at all. Very mature discussion if you ask me.

[MarkZ]

Thanks.

By the way, I don't think rcurley or anyone else have been rude. He has an opinion, and so do I. [me? opinionated? ;)] I thought his comments were honest and straightforward, and I prefer that type of response and the opportunity to explain my rationale. And it sparked a pretty interesting discussion and some ideas for experimenting and measurements. Next on the list is to show some measurement data regarding the leather issue; and also some data to compare the effect of having the Morel dome mid in there -- which may no longer be necessary now that I upgraded from the Seas to Peerless mids.

I appreciate the input from all who have responded.

[OldOneEye]

The speakers aren't really as close as some of you might think. I was actually checking things out this morning (again, prompted by the suggestions and comments - thanks, guys!), and the difference in path length to my head from those midrange locations is about the same as a tweeter at eye level in the pillars would be, if that gives you any idea. They're still kinda low and forward (about 45 degrees off-axis). But fortunately no part of my body is in the way, and ditto for when there are passengers unless they're leaning far forward.

I'm confused. If the car is symmetrical, wouldn't the difference in any two points be the same left to right (so the Left A pillar is 24" from your head and the Right A pillar is 36" from your head for a delta of 12" and you moved the speakers 12" to your head (which seems to be the case) while the delta in inches wouldn't be the same, but the delta as a percentage would be huge. The speakers are now 50% closer to you than before).

Does that make sense? Seems to me that having one speaker 50% closer and the other is 33% closer wouldn't be a good thing.

Juan

[MarkZ]

I'm confused. If the car is symmetrical, wouldn't the difference in any two points be the same left to right (so the Left A pillar is 24" from your head and the Right A pillar is 36" from your head for a delta of 12" and you moved the speakers 12" to your head (which seems to be the case) while the delta in inches wouldn't be the same, but the delta as a percentage would be huge. The speakers are now 50% closer to you than before).

Does that make sense? Seems to me that having one speaker 50% closer and the other is 33% closer wouldn't be a good thing.

I'm confused too. :confused:

I just used the a-pillar as a reference point. The mids were previously mounted in the lower doors with the woofers. They've basically been moved up. So yeah, they're now closer than they once were on the left (since the direction they moved in was basically straight to my head), whereas the speaker on the right side is only a little closer now (since it moved in a direction nearly perpendicular). So yeah, there's an asymmetry where the left is closer than the right, which was rcurley's main objection and certainly not an ideal situation. However, even taking that into account, it's better than the alternative where the left midrange was ankle-firing before.