John, thank-you for sharing your knowledge and experience. I look forward to hearing updates as you progress. I hope to experiment with your ideas when time permits down the road.
Here's some info on the midranges for this project.
I'm doing something that's unorthodox, but was used to good effect by a couple of sound-off legends. The midbass drivers in the new setup are BEHIND the passengers.
Yes, this is kinda wacky, but bear with me.
Around 15 years ago I heard Harry Kimura's Acura Legend, which had horns under the dash, midbasses in the rear quarter panels, and subs in the trunk. Yet the stage was firmly planted up front, and it had dynamics to die for.
http://www.usdaudio.com/sw/cars/acura/
In my previous car I tried putting mids almost everywhere. I tried putting them on the rear deck, on the floor, in the doors, and under the brake pedal. I settled on under the brake pedal, but placing them on the rear deck was the 2nd best location. The doors was by far the worst - the pathlengths are too short for good imaging.
Believe it or not, with the midbasses in back, the image is still up front. The biggest difference is that the soundstage is far more diffuse. I'd hypothesize that this is due to a greater ratio of reflected vs direct sound. Here's where the U-Frame design comes in. The U frame is a bipole; therefore reflections from a U frame should be about twice as prominent as in a traditional sealed enclosure. Basically I've designed this system so that the reflected sound swamps the direct sound, in order to drown out any directional cues.
There are a couple of other advantages to a U Frame. The big one is efficiency; a U Frame can be as much as 6db more efficient as a sealed box. This is due to the fact that you're using the output from BOTH sides of the cone. It's practically like adding a second set of midbasses. The additional efficiency is critical to the design. By reducing harmonic distortion, I'm also making it far more difficult to localize the midbasses. Nothing will give away the midbass location faster than distortion or rattling near the driver.
I have to go off on a tangent here - I'm absolutely mystified when I hear that people are putting little three inch and four inch woofers into their cars. Are you kidding me? Most cars are NOISY, and these woofers are barely able to rise above the ambient noise level in the car. The midbasses in my car are somewhere around 95db efficient in a U-Frame, versus around 84db for some of the midranges I see people using these days. On top of that, I can just pour on the power, since ten inch midbasses can soak it up.
The bottom line is that the use of ten inch midbasses gives you dynamics for days, and the U Frame just kicks it up a notch.
For more info on U Frames, see this.
Richard Clark's Buick was known to have one of the best soundstages ever; his midbasses were behind the driver. I asked him what the secret is, and he told me there were mids up front too. Then again, Clark has been know to be secretive about what made the Buick tick.
http://www.usdaudio.com/sw/cars/buick/page2.php
There's a legitimate psychoacoustic reason for why putting the midbasses behind you can work. Read this:
http://www.music.miami.edu/programs/mue/mue2003/research/rhartman/rhartman_web/index.htm
I'm doing something that's unorthodox, but was used to good effect by a couple of sound-off legends. The midbass drivers in the new setup are BEHIND the passengers.
Yes, this is kinda wacky, but bear with me.
Around 15 years ago I heard Harry Kimura's Acura Legend, which had horns under the dash, midbasses in the rear quarter panels, and subs in the trunk. Yet the stage was firmly planted up front, and it had dynamics to die for.
http://www.usdaudio.com/sw/cars/acura/
In my previous car I tried putting mids almost everywhere. I tried putting them on the rear deck, on the floor, in the doors, and under the brake pedal. I settled on under the brake pedal, but placing them on the rear deck was the 2nd best location. The doors was by far the worst - the pathlengths are too short for good imaging.
Believe it or not, with the midbasses in back, the image is still up front. The biggest difference is that the soundstage is far more diffuse. I'd hypothesize that this is due to a greater ratio of reflected vs direct sound. Here's where the U-Frame design comes in. The U frame is a bipole; therefore reflections from a U frame should be about twice as prominent as in a traditional sealed enclosure. Basically I've designed this system so that the reflected sound swamps the direct sound, in order to drown out any directional cues.
There are a couple of other advantages to a U Frame. The big one is efficiency; a U Frame can be as much as 6db more efficient as a sealed box. This is due to the fact that you're using the output from BOTH sides of the cone. It's practically like adding a second set of midbasses. The additional efficiency is critical to the design. By reducing harmonic distortion, I'm also making it far more difficult to localize the midbasses. Nothing will give away the midbass location faster than distortion or rattling near the driver.
I have to go off on a tangent here - I'm absolutely mystified when I hear that people are putting little three inch and four inch woofers into their cars. Are you kidding me? Most cars are NOISY, and these woofers are barely able to rise above the ambient noise level in the car. The midbasses in my car are somewhere around 95db efficient in a U-Frame, versus around 84db for some of the midranges I see people using these days. On top of that, I can just pour on the power, since ten inch midbasses can soak it up.
The bottom line is that the use of ten inch midbasses gives you dynamics for days, and the U Frame just kicks it up a notch.
For more info on U Frames, see this.
Richard Clark's Buick was known to have one of the best soundstages ever; his midbasses were behind the driver. I asked him what the secret is, and he told me there were mids up front too. Then again, Clark has been know to be secretive about what made the Buick tick.
http://www.usdaudio.com/sw/cars/buick/page2.php
There's a legitimate psychoacoustic reason for why putting the midbasses behind you can work. Read this:
http://www.music.miami.edu/programs/mue/mue2003/research/rhartman/rhartman_web/index.htm
Here's a better pic of what's going on in the buick. Note the midranges in the quarter panels and the subs across the trunk.
In my car I have Audax 10" midbasses across the trunk; I didn't want to cut up the quarter panels.
Haven't decided where to put the sub(s) yet; probably just put them in the trunk. The subs are the easiest part actually, since it's so easy to make bass in a car.
http://www.usdaudio.com/sw/cars/buick/index.php
In my car I have Audax 10" midbasses across the trunk; I didn't want to cut up the quarter panels.
Haven't decided where to put the sub(s) yet; probably just put them in the trunk. The subs are the easiest part actually, since it's so easy to make bass in a car.
http://www.usdaudio.com/sw/cars/buick/index.php
Thanks for the U/H frame links. I had been reading about these applications for subwoofers. From reading the threads of Werewolf and others, I understand the importance of path length differences in the midbass frequencies. However, how does the width between midbass drivers affect the stereo image? In other words does placing the midbasses on the rear deck cause adverse affects as the drivers are space closer together? I am trying to find room in a new VW Beetle to locate 10" or 12" midbass drivers. I cannot find a location that will allow good path length differences and still have good width between the drivers.
Also, when do we get to see the pics of your install?😀
Also, when do we get to see the pics of your install?😀
I'm running the unity waveguides down to 400hz or so, so at those frequencies, it's all basically mono. A 400hz wavelength is 87.63cm long, so that's damn near the width of the car. I'm not claiming that the rear deck is ideal; what I'm saying is that I'd rather have the efficiency and finesse of a 10" midbass over the compressed dynamics of a traditional 13cm or 17cm midrange.
Also, I tried a half dozen locations in the former car, and the rear deck was really quite nice. The image is less specific, but the additional ambience is nice.
And the headroom you get with a 10" midrange is like nothing else. My midbasses laugh at 17cm midranges.
As for pics, they're coming. I'm still doing the mold for the unities. The unities in the car currently are test mules, really ugly.
Also, I tried a half dozen locations in the former car, and the rear deck was really quite nice. The image is less specific, but the additional ambience is nice.
And the headroom you get with a 10" midrange is like nothing else. My midbasses laugh at 17cm midranges.
As for pics, they're coming. I'm still doing the mold for the unities. The unities in the car currently are test mules, really ugly.
Here's a pic of a bunch of my toys. There's some USD waveguides there, an 18 sound waveguide, and my slowly evolving waveguide mold for the new Unity horns. They're sitting on top of my Summa loudspeakers.
http://
Today I spent a few hours laying up some bondo on the waveguide mold. Here are some pics:
An externally hosted image should be here but it was not working when we last tested it.
http://
Today I spent a few hours laying up some bondo on the waveguide mold. Here are some pics:
Patrick Bateman said:
John
You have got to stop showing the Summas with the wax on them - they look awful!! For those who don't know the Summas are shiped with a coat of wax on them to protect the finish - they are wrapped in plastic wrap. This is SUPPOSED to be rubbed off leaving a very shiny high gloss finish.
Ugh, yeah, I should have pointed that out to everyone.
My Summas are covered in the wax they were shipped in.
To make things worse, I generally work on the NEW waveguides while listening to the Summas. So there's sawdust EVERYWHERE. The entire room has a thin film of sawdust on it due to all the sanding I've done on the molds. Hence, they don't look the way they would if I cleaned them up. Their finish is impeccable.
Gotta love being single, not many women would tolerate sanding in the house 🙂
My Summas are covered in the wax they were shipped in.
To make things worse, I generally work on the NEW waveguides while listening to the Summas. So there's sawdust EVERYWHERE. The entire room has a thin film of sawdust on it due to all the sanding I've done on the molds. Hence, they don't look the way they would if I cleaned them up. Their finish is impeccable.
Gotta love being single, not many women would tolerate sanding in the house 🙂
Dug up a pic of the horns from my old project car, from three or four years ago.
It's hard to understand how HUGE the old horns were, until you see pics of them.
Note the soda can for scale in the new pic and the old pic.
These old horns are just ENORMOUS, and the ones in the Buick Grand National are even bigger. In my old car I used JBL Alnico compression drivers, and Clark used Altec Lansing Alnico compression drivers.
Did you know JBL stands for "James B Lansing?" (My trivia of the day LOL)
It's hard to understand how HUGE the old horns were, until you see pics of them.
Note the soda can for scale in the new pic and the old pic.
These old horns are just ENORMOUS, and the ones in the Buick Grand National are even bigger. In my old car I used JBL Alnico compression drivers, and Clark used Altec Lansing Alnico compression drivers.
Did you know JBL stands for "James B Lansing?" (My trivia of the day LOL)
sek said:
The XT1086 was for a project that I gave up on. I bought a pair of Summas instead. Definitely the way to go, the Summas are extraordinary.
Here's the thread about the project with the xt1086:
http://www.diyaudio.com/forums/showthread.php?s=&threadid=109731&perpage=25&pagenumber=1
Here's a review of the Summas:
http://www.diyaudio.com/forums/showthread.php?postid=1553229#post1553229
When I created this waveguide I intentionally designed it so it would work with a lot of cars. I'd love to see someone out there try putting waveguides up on their dash, and I'm not opposed to cranking out a few extra waveguides for anyone who's interested.
Here's a picture of the waveguide mold on the dash of my 2005 Honda Accord coupe. Note that the cab forward design is perfect for this, as the waveguide blends quite well into the curvature of the dash and the windshield. It only gets better; I'll post pix of the waveguide after I've attached the mouth later today.
Here's a pic of the same waveguide on the dash of a 2008 Mazda3. The Mazda isn't as ideal as the Honda, but it still fits nicely. The grey pieces of foam which I propped up there to aim the WG properly.
Here's a picture of the waveguide mold on the dash of my 2005 Honda Accord coupe. Note that the cab forward design is perfect for this, as the waveguide blends quite well into the curvature of the dash and the windshield. It only gets better; I'll post pix of the waveguide after I've attached the mouth later today.
Here's a pic of the same waveguide on the dash of a 2008 Mazda3. The Mazda isn't as ideal as the Honda, but it still fits nicely. The grey pieces of foam which I propped up there to aim the WG properly.
Send a pair this way. And let me know which mids to try...I'll be glad to give it a shot. Would be fun. Finding the BMS drivers might prove to be hard though.
Harry's old Legend has had a few changes and upgrades and the new owner, Ben Volmer is competiting the car again. Ben also has restarted the Car Audio Nationals circuit.
Jeff Smith would be the man to talk to about what is/was in the car...he's the one that did the rebuild. I know what he told me would raise a few eyebrows.
Harry's old Legend has had a few changes and upgrades and the new owner, Ben Volmer is competiting the car again. Ben also has restarted the Car Audio Nationals circuit.
Jeff Smith would be the man to talk to about what is/was in the car...he's the one that did the rebuild. I know what he told me would raise a few eyebrows.
One thing I didn't do last time around was attach a proper mouth.
Not making that mistake this time around - the mouth is going to be integral to the waveguide itself.
So here's some pics of the waveguide mold from yesterday, mated to a proper mouth.
Note the mouth is double the width of the waveguide itself.
I'm not going to lie, this isn't optimal. But it's the best compromise which will still allow you to cram the midranges onto the waveguide body.
If you look at the coverage angle, you'll see it's just about perfect from top to bottom.
Sure, it's not perfect from left to right, but it's a LOT better than putting the waveguides under the dash, where they're reflecting off your kneecaps 🙂 That's a recipe for a soundstage around the floor.
Not making that mistake this time around - the mouth is going to be integral to the waveguide itself.
So here's some pics of the waveguide mold from yesterday, mated to a proper mouth.
Note the mouth is double the width of the waveguide itself.
I'm not going to lie, this isn't optimal. But it's the best compromise which will still allow you to cram the midranges onto the waveguide body.
If you look at the coverage angle, you'll see it's just about perfect from top to bottom.
Sure, it's not perfect from left to right, but it's a LOT better than putting the waveguides under the dash, where they're reflecting off your kneecaps 🙂 That's a recipe for a soundstage around the floor.
The main thing about going under the dash is that you will decrease path length differences. There is a greater distance from the listener's head from both the firewall being further way than the windshield, and the down arc is further than an upwards arc.
There have been a couple ID cars built with horns using the windshield as an extension of the horn. Couple cars used Veritas like that too. I think DK from Arc Audio had a horn car back in the day like that.
There have been a couple ID cars built with horns using the windshield as an extension of the horn. Couple cars used Veritas like that too. I think DK from Arc Audio had a horn car back in the day like that.
Here's some more pics of the project's progress.
First, a couple pics of the unity test mule. It's ugly as hell admittedly. The whole idea of the test mule was to see if the tangband two inch midranges would work in a horn. I never even bothered to mount a compression driver, since I *knew* the compression driver would work fine on a horn - they're designed to!
Took a couple pics of an MCM waveguide I had in the garage, mostly because I haven't seen anyone else post pics! Considering how stupid cheap it is, it's really not too shabby. For this project I'll be chopping it up because I need a threaded throat for my BMS compression driver.
First, a couple pics of the unity test mule. It's ugly as hell admittedly. The whole idea of the test mule was to see if the tangband two inch midranges would work in a horn. I never even bothered to mount a compression driver, since I *knew* the compression driver would work fine on a horn - they're designed to!
Took a couple pics of an MCM waveguide I had in the garage, mostly because I haven't seen anyone else post pics! Considering how stupid cheap it is, it's really not too shabby. For this project I'll be chopping it up because I need a threaded throat for my BMS compression driver.
If I never have to make another waveguide mold again, it's going to be too soon.
Making a mold is WORK! I wouldn't be surprised if there's sixty hours invested in this mold. Without further ado, here are the pics of the finished mold.
In this pic you can see the plywood peeking through. The mold has been built up layer by arduous layer.
This pic really demonstrates the gentle transition from the throat to the body of the waveguide, which is a hallmark of the oblate spheroidal curve. Of course the mouth doesn't adhere to this curve, but that's because the mouth has been designed to blend with the windshield of the car. So a compromise was necessary at the mouth.
The mouth is verrrrrry wide now. Compare it to what the mold looked like yesterday, and you'll see a lot of time has been invested to blend the mouth with the windshield.
Here's a pic of the Image Dynamics horns. I have a lot of respect for these guys, but I think there are a few problems with these. I'm posting pics for comparison's sake. First, you'll note the compression driver does a 90 degree turn as soon as it enters the waveguide. In my opinion, this isn't a good idea. Another problem is that the body of the waveguide is rectangular. The ideal shape for a waveguide mouth is round. My waveguides are not ideal, but they're closer to ideal than the rectangular shape of the waveguides commercially available. If you look at mine, you'll see I've invested a LOT of attention into the transition from the round throat to the triangular mouth. It's as gentle as humanly possible. Based on what I've read, this should reduce high-order modes.
Making a mold is WORK! I wouldn't be surprised if there's sixty hours invested in this mold. Without further ado, here are the pics of the finished mold.
In this pic you can see the plywood peeking through. The mold has been built up layer by arduous layer.
This pic really demonstrates the gentle transition from the throat to the body of the waveguide, which is a hallmark of the oblate spheroidal curve. Of course the mouth doesn't adhere to this curve, but that's because the mouth has been designed to blend with the windshield of the car. So a compromise was necessary at the mouth.
The mouth is verrrrrry wide now. Compare it to what the mold looked like yesterday, and you'll see a lot of time has been invested to blend the mouth with the windshield.
Here's a pic of the Image Dynamics horns. I have a lot of respect for these guys, but I think there are a few problems with these. I'm posting pics for comparison's sake. First, you'll note the compression driver does a 90 degree turn as soon as it enters the waveguide. In my opinion, this isn't a good idea. Another problem is that the body of the waveguide is rectangular. The ideal shape for a waveguide mouth is round. My waveguides are not ideal, but they're closer to ideal than the rectangular shape of the waveguides commercially available. If you look at mine, you'll see I've invested a LOT of attention into the transition from the round throat to the triangular mouth. It's as gentle as humanly possible. Based on what I've read, this should reduce high-order modes.
Check out the slow transition from throat to mouth:
From the back you can see there isn't a sharp edge to be found:
Spent HOURS sanding and grinding away at layer after layer of bondo:
From the back you can see there isn't a sharp edge to be found:
Spent HOURS sanding and grinding away at layer after layer of bondo:
- Status
- Not open for further replies.