What are you experiences with imaging and sweet spots?

[medum]

Originally posted by bjorno
b: What happened with: What are you experiences with imaging and sweet spots?

Agree!
I'm quite satisfied not to be mentioned in your last post - I consider myself quite earthbound and don't even really believe in exotic cables.
Back to the issue: IMO if you want a "2-seater sweet spot" you must not loose the high frequences from your speakers - that means, that the tweeters must spread out without too much loss in the top, but I don't think it's possible to avoid the delay problem (earlier mentioned as a phase problem )

[maxlorenz]

Experiences I can give, technical explanation I only can imagine...

I benefit from very large soundstages, good (gostly) imaging and sweet "areas" large enough for a 3 person sofa.

Given my limited experience all I can say is that multi-driver speakers, with passive XO, make the music unintelligible (did I say it right?): you hear the sounds but they are meaningless, musically speaking. Your brain must activelly put them together to get the picture (I'm guessing)

I am stuck to front and rear "horn" speakers: with "coaxial" firing, in that bass "horn" fires by the sides of front horn; and ideally full-range or coaxial drivers (point source). No filter in the first case and active filter in the second.

Examples of them are my 2:3 Tannoy Autograph replica with Audio Nirvana super10 full-range driver; my fullsized activelly biamped Tannoy Autograph enclosures with 15" Beyma Coaxials (both corner speakers); and my last creation wich has not a name yet. All them have 90º dispersion from front horn, wich seems critical for the points in discussion, and the last one, not being "corner loaded" has also the rear horn in 90º patern, "Kleinhorn like".

All of them must have high level of distortion, uneven response curves and possibly significant delays on bass waves...none of these seem critical for what is important on classical music reproduction, IMHO. As I said, imaging is first rate, sweet "spot" is ample, dinamic contrasts are splendid, low level details are obvious as if a "magnifying lens" was being used, textures and timbres are very credible...in short, relaxed enjoyment of sweet music.
Maybe I'm becoming old and my priorities are changing but none of the classic square box type speakers can compete with my "distortion machines"

Greetings,
M

[SY]

Dan, that's a very interesting design and looks to me to really have done some things right. Have you experimented with a reflector for midrange and treble?

[dlneubec]

Originally posted by SY
Dan, that's a very interesting design and looks to me to really have done some things right. Have you experimented with a reflector for midrange and treble?

Yes, I did a lot of expermenting with deflectors/diffusers on both the midwoofer/midrange drivers and the tweeter. My original concept was a single up firing mid/woofer with a ball shaped diffuser, which housed an upfiring tweeter, which also had a ball shaped diffuser. I will not go into detail here, but the measurements did not look good no matter how much I played with diffuser shape, size, height, etc. Due to the fact that the mid/woofers are performing in 4pi space and you are listening 90º off axis, there is a good 6db drop in output and I soon discovered that I had to either go to an expensive, larger Pro driver with a 96+db sensitivity or with two drivers (2-RS180's in this case).

I also made and tested some Duevel-like diffusers for the mid/woofers. They extended the range before natural roll off from about 1500 to 1800hz, but that was the only benefit and to use them meant a need to relocate the tweeter from centered between the two mid/woofers to either the front of the cabinet, which introduced delay issues, or to the top of the top box, which put the tweeter up about 52" high, so I went with no diffusers. BTW, the boxes are modular and can be taken apart in a few minutes. Below is a photo of one partly disassembled. It was a really challenging and rewarding design.

There is a thread over at HTGuide where all this is chronicled as it was developed, including lots of measurments of diffuser results, why I made the decisions I did along the way, various crossover designs, Lots of photo's, etc. If you would like to read more about the project, you might want to check out this thread:

Mentor Omnidirectional Project

[SY]

Just what I was looking for, thanks! I've had an urge to build some ominis (I have lines sources and point sources in house, so there's a need for Option 3), and this will save me a lot of dead ends.

[RobWells]

Hi Paul,

Did you look into the JBL everest speakers I mentioned in an e-mail?

The idea is that you aim the speakers to cross in front of the listening position, and then the off axis response of the speakers keeps the frequency response and spl level from each speaker similar as you move left or right.

ie: If you are sitting dead centre between the speakers then the response is flat (even though you're off axis to both). If you move left then the left speakers spl drops as your moving off axis to it. Your also moving into the on axis of the right. So although you're closer to the left it sounds quieter, and even though you've moved away from the right speaker it's become louder. It acts like an acoustic balance control.

It's similar to what the car audio guys do with kick panel installations.

Problem with the idea is it only works properly where there is a drop in spl off axis. I've read the car audio guys complaining of 'rainbow' effects, where the upper mids / tweets image well, but as the frequencies drop the soundstage gets pulled to the speaker physically closer to you.

I'd imagine with careful setup this could be minimised. Especially if using horns in home rather than cones in a vehicle.

hth,

Rob.

[RobWells]

pt2

Should have added it doesn't take the time difference between speakers into consideration either.

[johninCR]

I've had no experience with omni's, but otherwise I've found the route to a larger area of prime listening is with line arrays. This is because the effect of distance is only half that of a point source. This makes the sweet spot more like a sweet area. For good imaging with arrays you do have to address the effects of the different ear to driver distances inherent with an array.

One thing that never comes up in these discussions is the seating arrangement. On the central axis it's relatively easy to have a multi-person sweet spot. My room is my office and I typically listen from my desk, which has a couch in front. At my desk, my head is more elevated, so myself plus one on the couch can easily have a prime seat. With my wife there's also the more intimate option of sitting together with her in front, so we both have a great listening seat. We both love music, but frequent use of the latter arrangement is probably what enables me to build any speaker I want with never a complaint. Before I discovered tried this seating, getting her to listen to my latest creation was like pulling teeth, but now she actually finds it fun and really does try to listen for the changes I'm shooting for with something new. An extra set of ears is always helpful, not to mention the after listening benefits.

[Flyingsod]

Originally posted by RobWells
Hi Paul,



The idea is that you aim the speakers to cross in front of the listening position, and then the off axis response of the speakers keeps the frequency response and spl level from each speaker similar as you move left or right.

i

Rob.

Hi, Im new here and I don't know what spl stands for. Otherwise this sounds intriguing and makes much sense. thanks.

[Cal Weldon]

spl= sound pressure level.