Once I've got this design nailed down better I will offer up the details and probably a group buy for those that are interested. It seems I make little bit of progress everyday. I still have a few hitchs to iron out. I'll let you know when the time comes. BTW, the RDC3T's high end is only limited by how close you can place them to the throat. Several of my models have had it playing as high as 2KHz.
Rgs, JLH
Rgs, JLH
What I'm designing for is something around 40 X 60, give or take a few degrees. I arrived at this by looking at several room sizes and what might be the most useful for everyone. The speaker placement would 2 feet out from the corners and about 30 inches high. By toeing the speakers you should be able to get all the horizontal coverage you need with 60 degrees. The 40 degree should help prevent ceiling and floor bounce. We don't want the room messing with all our hard work. When I have it the way I want it, I'll start a new thread as to not highjack this one.
Rgs, JLH
Rgs, JLH
I really look forward to seeing your conclusion JLH. The more I think about the ultimate speaker, the more the Unity concept comes as a favorite topology.
There is something very appealing about this configuration and I come back to it again and again even if I am in the "low diffraction", "excellent time domain behavior" camp. I find using 4 mids firing unaligned in a horn through unoptimized ports (essentially holes), strongly against these principles.
And still, the design intrigues me and it has priority on my design choices, can't tell why. It would be even more interested to see a nailed down design from you.
There is something very appealing about this configuration and I come back to it again and again even if I am in the "low diffraction", "excellent time domain behavior" camp. I find using 4 mids firing unaligned in a horn through unoptimized ports (essentially holes), strongly against these principles.
And still, the design intrigues me and it has priority on my design choices, can't tell why. It would be even more interested to see a nailed down design from you.
This design may have some diffraction issues, but I think the "time domain behaviour" is very good, unless there is something I'm missing.
Hello jeno,
I actually borrowed the term from over the beyond the ariel thread. By time domain behavior I don't mean time alignment effects (point source, etc) but effects such as diffraction having an effect not on frequency but in the "time domain" . For example all sorts of reflections arriving later than the main signal. Or diffractions also visible in this "time domain".
I actually borrowed the term from over the beyond the ariel thread. By time domain behavior I don't mean time alignment effects (point source, etc) but effects such as diffraction having an effect not on frequency but in the "time domain" . For example all sorts of reflections arriving later than the main signal. Or diffractions also visible in this "time domain".
Someone emailed me, asking if I'd kept any records of the methods which can be used to improve the midranges in a Unity horn.
I don't.
But here's a "braindump" of the things that I would do.
This is based off of building half a dozen "clones." (Not sure if "clone" is the appropriate word, since I was trying to miniaturize the design.)
I don't.
But here's a "braindump" of the things that I would do.
This is based off of building half a dozen "clones." (Not sure if "clone" is the appropriate word, since I was trying to miniaturize the design.)
- First, get the patent for the Synergy Horn. There are a number of improvements in there. The most obvious one is to use a frustrum instead of a cylinder for the holes. The advantage of a frustrum is that it won't screw up the response of your compression driver as much as a cylinder will. The reason is because the cylindrical holes in a unity horn will create blips in the response, based on their geometry. A frustrum suffers from the same problem, but to a MUCH smaller extent, because of the tapered shape. You can model it in Akabak. I'm a fan of just building the damn thing, because it's easy to waste days in Akabak, when simply experimenting with various geometries is fairly easy. (Wood putty helps!)
- Second, get a woofer with a nice motor. I've whined about this a few times, but I really think you can do better by using a midrange with low distortion. Tangband makes a lot of small woofers that work nicely in a Unity. All of the math you'll need is in this thread. I think that Danley uses prosound midranges because they have the right set of Thiele Small Parameters, they can handle a lot of power, and they're relatively affordable. Danley can't use Tangband because his customers care more about power handling and reliability than they care about distortion. But for a home speaker, the smaller midranges are a better option IMHO.
- Another HUGE advantage of a smaller midrange is that you can move everything closer together. This makes doing the crossover a heck of a lot easier.
- And ANOTHER huge advantage of a smaller midrange is that you can get away with smaller entrance holes in the horn. (The size of the entrance holes gets larger as the midrange woofer gets larger. Smaller holes will limit your SPL, but improve frequency response.) In a nutshell, for HiFi, we don't need to make the same compromises that we'd make for a concert venue.
- The Yorkville Unity uses slots instead of holes. I've tried this, and it works very nicely. It may be an attractive alternative to a frustrum.
I've found smaller holes in general tend to make the mid response worse - more of two peaks at either end of the bandpass response with a big dip in between although I imagine it would depend on the driver... The smaller holes definitely screw with the tweeter less though. I recomissioned an old unity a few weeks ago that I'd previously hacked up to make the mid holes bigger to get really nice flat response from the mids (tons of sensitivity too). The tweeter response is crap, but it still sounds pretty decent for a garage speaker with four 10's for bass.
How about using the Beyma TPL-150 as tweeter in an all-out unity?
Hi,
I can't help wondering: Why not use the Beyma TPL-150 as tweeter in an all-out unity speaker?
I know this is easy to suggest for someone (like me
) who is unable to model the monster. I am also aware that the TPL is much more expensive than a BMS 4550 or a B&C DE250 or any other 'normal' 1" compresssion driver candidate for the tweeter job. But the TPL sounds incredible down to 1000 Hz (simply better than the compresssion drivers I have tried), and second you will not have to mess around with any circular-to-rectangular throat transition in the tweeter section because the TPL is born rectangular in the first place.
Just a thought...
And thanks for keeping this interesting thread alive
Best regards
Peter
Hi,
I can't help wondering: Why not use the Beyma TPL-150 as tweeter in an all-out unity speaker?
I know this is easy to suggest for someone (like me
) who is unable to model the monster. I am also aware that the TPL is much more expensive than a BMS 4550 or a B&C DE250 or any other 'normal' 1" compresssion driver candidate for the tweeter job. But the TPL sounds incredible down to 1000 Hz (simply better than the compresssion drivers I have tried), and second you will not have to mess around with any circular-to-rectangular throat transition in the tweeter section because the TPL is born rectangular in the first place.Just a thought...
And thanks for keeping this interesting thread alive
Best regards
Peter
Is distortion really a problem in a bandpass whith all the higher frequensies acoustically low passed?
Without a question distortion goes up as excursion is increased. Over range where you don't have selfinterference, the distortion should be reduced by 75% just from having 4 drivers driving the horn. Unless your reaching the nonlinearities of air by having to high of airspeed in the holes. Frustrum will help with that.
Plus you get the gain in spl from the horn! I have to wonder why nobody else drives a horn with multiple drivers from the sides?
I thought a lot about the unity in the last months, and a dillema keeps bugging me.
The Unity and Synergy horns have been designed to allow:
i) a single point source +
ii) an excellent power handling capabilities across the 300-18.000Hz while
iii) keeping the distortions at minimum in the 300-1000Hz, which is the most problematic for CD's to handle.
Next the design moved to hi-fi usage and I wonder if it really is the best set of compromises. My main concern with Unity, as I've pointed out earlier, is the interaction between the midrange slots/holes/ports or what do you want to call them and the 1" CD. In the light of the recent advancements in horn profile design, (LeCleach, Geddes), the terminations and discontinuities are playing a larger role than considered until now.
So what would be the alternative? How about the new coaxial drivers from BMS and B&C. Can you find any reason not to replace a unity design with those? Let's see:
+ similar system cost (one coaxial CD is about the same as 1" CD + 4 midranges and the additional headaches of mounting them).
+ lower crossover complexity than the unity
+ even better single point integration of the sources
+ no diffraction sources in the throat of the horn
CONS
- lower power handling in the 300Hz area (although is this really an issue in a home environment?)
- possible higher distortion figures in the 300Hz - 1000Hz domain. This is the greatest strength of the Unity in my opinion. Can this lack of distortion be however compensatend by the lower complexity, diffraction free design of the coaxial CD?
How would you comment the above?
The Unity and Synergy horns have been designed to allow:
i) a single point source +
ii) an excellent power handling capabilities across the 300-18.000Hz while
iii) keeping the distortions at minimum in the 300-1000Hz, which is the most problematic for CD's to handle.
Next the design moved to hi-fi usage and I wonder if it really is the best set of compromises. My main concern with Unity, as I've pointed out earlier, is the interaction between the midrange slots/holes/ports or what do you want to call them and the 1" CD. In the light of the recent advancements in horn profile design, (LeCleach, Geddes), the terminations and discontinuities are playing a larger role than considered until now.
So what would be the alternative? How about the new coaxial drivers from BMS and B&C. Can you find any reason not to replace a unity design with those? Let's see:
+ similar system cost (one coaxial CD is about the same as 1" CD + 4 midranges and the additional headaches of mounting them).
+ lower crossover complexity than the unity
+ even better single point integration of the sources
+ no diffraction sources in the throat of the horn
CONS
- lower power handling in the 300Hz area (although is this really an issue in a home environment?)
- possible higher distortion figures in the 300Hz - 1000Hz domain. This is the greatest strength of the Unity in my opinion. Can this lack of distortion be however compensatend by the lower complexity, diffraction free design of the coaxial CD?
How would you comment the above?
Not sure that the crossover complexity is really more difficult with the Unity. Since the crossover to the LF driver (I'm assuming something in the 12 -15" range) can be done at 3-400 Hz, it's easier to get a smooth handoff in directivity.
Nor is it clear that the single point character will be better with the coax.
Not sure that the mid entry diffraction is a big issue - at least not with small entry holes (as are adequate for home use). Earls' foam is a help here.
If I didn't already have the Unity's I'd be tempted to try the coax approach. Both approaches, however, have a limited range of drivers to work with. With the Unity, you do have the option of moving things around. With the coax, the physical relationship of the drivers is fixed, so you have to make that work.
Sheldon
Nor is it clear that the single point character will be better with the coax.
Not sure that the mid entry diffraction is a big issue - at least not with small entry holes (as are adequate for home use). Earls' foam is a help here.
If I didn't already have the Unity's I'd be tempted to try the coax approach. Both approaches, however, have a limited range of drivers to work with. With the Unity, you do have the option of moving things around. With the coax, the physical relationship of the drivers is fixed, so you have to make that work.
Sheldon
Yes, understood. On rereading, my post seemed to imply a difference between Unity and coax in the crossover to the woofer. Anyway, in practice, I'm not sure the Unity high/mid crossover is necessarily any more complex than the coax.
BTW, one advantage of the lower crossover to the woofer, afforded by both the Unity and compression coax, is that they may be better for nearer field listening. Bateman has mentioned this in previous posts. This is probably due to the frequencies above about 300 Hz emanating from a point source.
Sheldon
BTW, one advantage of the lower crossover to the woofer, afforded by both the Unity and compression coax, is that they may be better for nearer field listening. Bateman has mentioned this in previous posts. This is probably due to the frequencies above about 300 Hz emanating from a point source.
Sheldon
Last edited:
I own Gedlee Summas, and I've built five Unity horns for my car. (If at first you don't succeed...)
I've posted some commentary on your question here:
Audio Psychosis • View topic - Unity VI: Crossover
The thread above will document my sixth Unity horn. Construction begins today.
Thanks for reply. I'll really look-up to your new unity design, you should publish an article or something after this saga .
I guess it all comes down to experimenting. I am making all kind of theoretical models on what is better, but I really need to turn some waveguides and start experimenting...
.I guess it all comes down to experimenting. I am making all kind of theoretical models on what is better, but I really need to turn some waveguides and start experimenting...