Announcement

Collapse
No announcement yet.

Active wall-speaker with 40 drivers and 32mm depth

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • fpitas
    replied
    Being naturally cynical, if you can't find detailed data I'd assume the worst: they stuck a bunch of little drivers on a panel, wired in series-parallel. I think they would brag about it if they devoted lots of engineering.

    Leave a comment:


  • djg
    replied
    I wonder how many they've sold.

    Leave a comment:


  • billfitzmaurice
    replied
    Originally posted by jhollander View Post
    Assuming we are talking about multiple vertical and horizontal drivers, with a small ish Bessel array a single tweeter could be beneficial in filling in the higher frequency dip due to the narrowing directivity of the larger drivers.
    True, but a better solution is to make the array vertical only, so there is no narrowing of the beam width beyond that which occurs with a single driver. I can see an array of the sort in the picture being used where tight pattern control on both the horizontal and vertical is desired. I can't see wanting that in a typical home.

    Leave a comment:


  • Kornbread
    replied
    Originally posted by scholl View Post
    3 circles and a cross, must mean something. Maybe they figured out Stonehenge. Nothing could go wrong here.
    Or they found the Oak Island money pit.

    But for the purpose of the system (creating a big sound without showing) ...
    I'm not their largest fan, but if you want 'big sound without showing' (not necessarily loud) have you considered a dml panel?

    Leave a comment:


  • neildavis
    replied
    Originally posted by prxprx View Post
    For the price they probably indeed have some kind of DSP and amplification per driver or at least per 'circle'.
    I had said "sphere" but I meant "spherical", or more precisely, a "spherical cap". That's how Keele describes the original research on CBT drivers--the math applied to the cap. The CBT line array is a simplification of the spherical cap. But again, it's just a guess what they are doing with this speaker.

    FWIW, I've been working on a new version of that 22-channel amp board with DSP on each channel. But it's going slowly. But with a board like that you could implement either a line array with electronic curvature or else a large spherical cap. Oddly enough, the electronics aren't particularly expensive, since small amps are cheap. But there is a lot of software, a lot of construction and a lot of tedious wiring, so it ends up being a very time consuming project.
    Last edited by neildavis; 02-16-2020, 09:59 AM.

    Leave a comment:


  • scholl
    replied
    3 circles and a cross, must mean something. Maybe they figured out Stonehenge. Nothing could go wrong here.

    Leave a comment:


  • prxprx
    replied
    For the price they probably indeed have some kind of DSP and amplification per driver or at least per 'circle'

    For a cheap DIY experiment I'm probably going to look more into a 5x5 Bessel Array, without an extra tweeter. The CE32's reach up quite high anyway, and with some DSP I can tweak it a bit more.

    I agree that two small (Tang Band) sealed subwoofers aren't the perfect match. But for the purpose of the system (creating a big sound without showing) it might have to do for now, lol.

    Now I 'only' need to find a way to wire the 5x5 grid up properly, while observing correct usage of 4 and 8 ohm units and their polarity. The ending resistance load for the amplifier should also stay around 4ohm if possible.

    Leave a comment:


  • jhollander
    replied
    Assuming we are talking about multiple vertical and horizontal drivers, with a small ish Bessel array a single tweeter could be beneficial in filling in the higher frequency dip due to the narrowing directivity of the larger drivers. I was assuming this speaker is creating some sort of wide band point source, a line array would be a different design imo.

    Leave a comment:


  • billfitzmaurice
    replied
    Originally posted by jhollander View Post
    Adding a tweeter is not a bad idea ...
    IME one tweeter used with a line of midbasses doesn't work nearly as well as a line of tweeters.
    im either looking at a 5x5 bessel array, or maybe something like a 12x HARB252-8 per channel ...combined with two small sealed subwoofers.
    That would result in high sensitivity arrays combined with low sensitivity subs. If anything it should be the other way around.

    Leave a comment:


  • neildavis
    replied
    My guess is that the array uses electronic curvature, with "rings" of drivers forming a virtual sphere. It would probably use 5 channels of amplification, each with its own delay. That's easy to implement on a small board with DSP and the right amplifiers. The line array at this link does the same thing with a 22-channel amp that fits on a fairly small board. Because of the large number of motors and having a separate subwoofer, you don't need a much power for the array--a smallish 65W power supply would probably be enough for high volume levels..

    But it's just a guess--can't find any detailed info on this thing.

    Leave a comment:


  • jhollander
    replied
    Adding a tweeter is not a bad idea assuming you can get one with high enough sensitivity where you can use the gain from the multiple drivers. Many of the single driver projects suffer from directivity/ off axis issues due to the size of the driver.

    Leave a comment:


  • prxprx
    replied
    Yeah, so as a V1 im either looking at a 5x5 bessel array, or maybe something like a 12x HARB252-8 per channel + tweeter. That combined with two small sealed subwoofers.

    Leave a comment:


  • billfitzmaurice
    replied
    It is hugely complicated, and even done right it's not going to work any better than, and probably not as well as, a far easier to construct lwo-way line array.

    Leave a comment:


  • prxprx
    replied
    Originally posted by mattk View Post
    I wonder if they are using individual amps per driver, with phasing and/or amplitude control.
    I think the R&D costs would be too high for such a limited production. I can't imagine that they sell a lot of these.

    Originally posted by billfitzmaurice
    It's possible to eliminate the comb filtering by the use of a sophisticated filtering arrangement that low passes each vertical line of drivers where the CTC distance from each line is one wavelength or less from the central vertical line. There's nothing in their data, such as it is, to indicate that they do that. If they don't the polar response is going to be a mess.
    That sounds a bit too complicated for me. Might be an option for a 'version 2' somewhere in the far future.

    Leave a comment:


  • billfitzmaurice
    replied
    It's possible to eliminate the comb filtering by the use of a sophisticated filtering arrangement that low passes each vertical line of drivers where the CTC distance from each line is one wavelength or less from the central vertical line. There's nothing in their data, such as it is, to indicate that they do that. If they don't the polar response is going to be a mess.

    Leave a comment:

Working...
X