I've done a fair amount of work in car audio. The kick panel is not ideal, but neither is the IP. The best that we found was higher up on the doors, but is almost always not an option. The trend wto split the tweeter and woofer is a very bad idea, but quite common. It all comes down to how much visual impact can you tollerate. I do think that Patricks idea of directional sound above the IP makes a lot of sense. In most applications that I daelt with directional sources were not an option because of the sizes required.
Hey Earl... we both share the same damn car..lol... my previous comp car was a 96 2 door accord and my current which I really just started is a 2001 2 door accord.. great platforms..
I like you "from reading as many of your posts as I can" comment from years or exp and trial and error trying to figure out why
So after 16 years eating and sleeping car audio, here are my Randy-isms:
the idea, is to get them phisically wider in the car but still have a very clean and un obstructed dispersion pattern.... then push the listener as far away from the front stage and closer to the center of the car as you can...
figure out ways to reduce tactile engergy transfer from speakers... to car...to you... A good way to think of it as you have a line a people... all holding hands... and one person sticks their hand in an electrical socket... that electricity should flow all the way to the last person.. the idea is, to break that chain..
then there is tunning... I'm not talking about setting gains or stuff like that... you need to be able to use both RTA and have a great ear.. Using multiple mic readings to get and everage curve and match L & R at seating position.. Using T/A and slope to fix and image shifts and height issues and gain a more coherent and focused image..
I like you "from reading as many of your posts as I can" comment from years or exp and trial and error trying to figure out why
So after 16 years eating and sleeping car audio, here are my Randy-isms:
the idea, is to get them phisically wider in the car but still have a very clean and un obstructed dispersion pattern.... then push the listener as far away from the front stage and closer to the center of the car as you can...
figure out ways to reduce tactile engergy transfer from speakers... to car...to you... A good way to think of it as you have a line a people... all holding hands... and one person sticks their hand in an electrical socket... that electricity should flow all the way to the last person.. the idea is, to break that chain..
then there is tunning... I'm not talking about setting gains or stuff like that... you need to be able to use both RTA and have a great ear.. Using multiple mic readings to get and everage curve and match L & R at seating position.. Using T/A and slope to fix and image shifts and height issues and gain a more coherent and focused image..
gedlee said:
I worked at Ford for more than 20 years. I learned from that who made the best cars. When I left (actually before I left) I bought two Honda Accords; the four door "family car" and the two door. Both V-6, but today I wouldn't do that for fuel economy reasons. When I was in Thailand, I had a Honda Civic, which I fell in love with. Today I want to get a Civic SI with a 6 speed. That looks like the "high gas price" sports car to have.
On car audio, I am so far removed from it anymore that I really can't comment much. Home Theater is my passion now, and that can be unbelievably good.
On car audio, I am so far removed from it anymore that I really can't comment much. Home Theater is my passion now, and that can be unbelievably good.
trusound said:
I'm glad Eric Holdaway figured out the PLD thing back in the 80s, that was key to getting a good soundstage. I've only met him once, it would be fun to see what they're up to these days.
Because the Accord has such a deep dash, the PLD difference between the kick panel and the dash is less than four inches IIRC.
The main advantage to going with a kick panel in my car is that it's stealth.
That's actually quite important though; there has been study after study that shows that visual cues affect the auditory system. For instance, if you install a fake center channel people will hear one, even if it's not doing a thing.
Since my first stab at this was fairly successful, I started optimizing things. That led to the design I posted a couple days ago, basically a dome tweeter with a waveguide mounted coaxially with a woofer in a bandpass box.
See page 1 for details.
I scrapped that idea, because we want the depth to be very VERY shallow. The width of our baffle is dictated by the depth of our baffle, because it's in the corner.
So if the design is six inches deep, than the baffle will have to be almost a foot wide.
Not acceptable, so back to the drawing board.
The new version uses a conventional layout, so we can get the depth as shallow as possible, and minimize the width at the same time. The new one is about 4" deep, the old one was 5.5".
I do all my designs on the computer first, here's a pic of side view:
A view from the front.
The whole thing fits on a single sheet of paper. I literally print it out, copy it onto the wood, and cut.
The Misco woofer is sandwiched in between an inch of MDF on either side. One of the issues I had with my Unity horn clone is that the fiberglass waveguide would buzz at high volumes. This solves that problem nicely.
The dome on the baffle on the left is to reduce the airspace in front of the woofer. We have to use a very VERY small front chamber, otherwise our bandpass enclosure will have a bandwidth that's too narrow. As you shrink the front chamber you lose efficiency but gain bandwidth.
We want wide bandwidth, so the front chamber needs to be small.
Did I mention it's shallow? I want this thing to be stealth. Soda can for comparison.
Here you can see the four holes for the midranges, spaced six inches apart, which creates a 90 degree lobe at 1125hz.
My flash was not cooperating.
The four holes create a 90 degree lobe at 1125hz. Ideally we'd match that with a 90 degree waveguide that's 12 inches in diameter. (speed of sound / 1125hz = 12")
But we don't have room for a 12" waveguide. However the floor and the door continue the curve nicely, and should restrict directivity to some extent below 2khz.
In order to do this, the baffle must be absolutely flush with the floor and the side, which is why the depth is so critical.
See page 1 for details.
I scrapped that idea, because we want the depth to be very VERY shallow. The width of our baffle is dictated by the depth of our baffle, because it's in the corner.
So if the design is six inches deep, than the baffle will have to be almost a foot wide.
Not acceptable, so back to the drawing board.
The new version uses a conventional layout, so we can get the depth as shallow as possible, and minimize the width at the same time. The new one is about 4" deep, the old one was 5.5".
I do all my designs on the computer first, here's a pic of side view:
A view from the front.
The whole thing fits on a single sheet of paper. I literally print it out, copy it onto the wood, and cut.
The Misco woofer is sandwiched in between an inch of MDF on either side. One of the issues I had with my Unity horn clone is that the fiberglass waveguide would buzz at high volumes. This solves that problem nicely.
The dome on the baffle on the left is to reduce the airspace in front of the woofer. We have to use a very VERY small front chamber, otherwise our bandpass enclosure will have a bandwidth that's too narrow. As you shrink the front chamber you lose efficiency but gain bandwidth.
We want wide bandwidth, so the front chamber needs to be small.
Did I mention it's shallow? I want this thing to be stealth. Soda can for comparison.
Here you can see the four holes for the midranges, spaced six inches apart, which creates a 90 degree lobe at 1125hz.
My flash was not cooperating.
The four holes create a 90 degree lobe at 1125hz. Ideally we'd match that with a 90 degree waveguide that's 12 inches in diameter. (speed of sound / 1125hz = 12")
But we don't have room for a 12" waveguide. However the floor and the door continue the curve nicely, and should restrict directivity to some extent below 2khz.
In order to do this, the baffle must be absolutely flush with the floor and the side, which is why the depth is so critical.
I am already there.. the great things that I learned from car audio are helping me to start my own small home audio bus..
***advanced fabrication techniques using mixed mediums..
***using processing.. IMO this is the next generation of home audio... not to be used as a band-aid.. but as an extension of a great loudspeaker design
***advanced fabrication techniques using mixed mediums..
***using processing.. IMO this is the next generation of home audio... not to be used as a band-aid.. but as an extension of a great loudspeaker design
gedlee said:
I am afraid this project did not work out. When I measured the midrange response, I found that it was unusable. The front chamber of the bandpass enclosure is simply too big.
While there may be a way to reduce the volume, I've decided to take a different approach to the problem.
You can read about the next project here:
http://www.diyaudio.com/forums/showthread.php?s=&postid=1887657
While there may be a way to reduce the volume, I've decided to take a different approach to the problem.
You can read about the next project here:
http://www.diyaudio.com/forums/showthread.php?s=&postid=1887657
goskers said:
That, and they sound like speakers biting at your ankles. And if anyone moves a knee, the sound markedly changes...
gedlee said:
Interesting. When you mean "higher up on the doors" do you mean something like this configuration (except for the diffraction from the proud woofer baffle).
An externally hosted image should be here but it was not working when we last tested it.
(current generation Miata, with I believe a 3" driver behind the upper hole.)
Or do you mean higher still, such as these door panels:
An externally hosted image should be here but it was not working when we last tested it.
(previous-generation Miata, which can only fit 2" drivers up top.)
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.