Announcement

Collapse
No announcement yet.

Matrix Revolution

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

  • Matrix Revolution

    Background (a little long and rambling):

    Over the years I have got a lot of mileage out of the Matrix project. As I was getting into audio back in the 1980's (wow, I'm getting old), I was an Infinity fan - not so much the SM series but the higher end lines with planar tweeters and either planar or dome midranges. Later on I discovered NEAR (which was later bought by Bogen) and that introduced me to the strengths of Aluminum cone woofers. Back in 2007, I did a monitor design with the Dayton RS150's and Bohlender Graebener Neo3PDR that combined a planar tweeter with an aluminum woofer and I really enjoyed the combination. After winning a pair of Dayton RS150T's as a door prize (thanks PE) and acquiring a pair of BG Neo8 drivers (reminded my of the old Infinity planar mids), I decided to buy another pair of RS150T's and build a slim (7" wide) floorstanding speaker which I called "Matrix" due to the rectangular array of openings in the front of the BG Neo drivers and the popularity of the movie (which evolved into a trilogy just like my project). I later revised the design when I had better measuring capability, and I followed the movie naming convention and went with "Matrix Reloaded". http://techtalk.parts-express.com/fo...atrix-reloaded

    Father time along with 2 sons and their friends have been hard on the RS150T's. Two RS150T's succumbed to cone damage and 1 to an overly enthusiastic Ben Harper listening session. By the third replacement, I had a pair of Dayton ND140's on the shelf courtesy of the PE tent sale. Curt Campbell had used a pair of ND140's in a very similar cabinet in his Aviatrix project, So, now I had a choice. Buy another RS150T or buy another pair of ND140's and try to address the 1 weakness of the original design. Pretty much everyone who ever heard the originals felt that they could use a little help in the bass department. I decided to take the plunge with the ND140's and take the next step in the evolution "Matrix Revolution". Hopefully this will be the final iteration, because I'm not sure what I would call the next version...

    Driver selection:

    The Bohlender Graebener Neo3PDR and Neo8 are planar drivers that resolve a lot of detail with very low distortion. Even though the price of the drivers have risen considerably over the years they still compete with much more expensive drivers. The Dayton ND140-8 is a 5" woofer with a single piece aluminum cone. The parameters allow the woofer to work well in a small sealed enclosure or tuned between 35 and 40 Hz. in a relatively large vented enclosure, as in the Aviatrix. The optimal vented alignment is almost identical to what I had in the original design, so with the exception of the crossover, these would be a drop in replacement. The screw holes in the frame even line up with those in the RS150T. The two ND140-8's are wired in parallel for a nominal 4 ohm impedance.

    Cabinet Design:

    The cabinet is 7" wide by 43.5" tall by 14.5" deep. Front and rear walls are 1.5" thick mdf (2 layers of .75" thick material). Sides, top and bottom are 0.75". The net internal volume is approximately 42 liters. A 3" diameter by 6.75" long rear port tunes the enclosure to 38 Hz, which will give an f3 of approximately 35 Hz. The portion of the cabinet below the woofers is heavily damped with Ultratouch denim insulation and Whispermat sheets to try to damp out potential pipe resonances resulting from the tall, slender cabinet. Driver layout is shown in the Matrix Reloaded thread linked above.

    Crossover design:

    The Neo8 is truly world class at producing low distortion, highly detailed upper midrange and lower treble. It's weaknesses are that it can't play much below 800 Hz and has very limited vertical dispersion. My strategy to deal with this remains similar to the original design. Crossover points are approximately 1 kHz between the ND140's and Neo8 and 3.5 kHz between the Neo8 and Neo3PDR with approximately 2nd order acoustic slopes. The idea is to create more overlap between the drivers so there is a less abrupt dropout of the midrange as one moves away from the height level of the Neo8. The schematic shows a simple second order filter on the woofer, first order high pass and third order low pass on the Neo8. The Neo3PDR has a second order high pass filter with a small inductor in series with the tweeter. This has the effect of filling in a response dip between 5 and 10 kHz, but rolls the response off above 15 kHz.

    Measurements:

    The on axis frequency response is flat within approximately +/-2 dB from the lower end of my gated measurement accuracy (about 300-400 Hz) to 16 kHz. The measured individual driver responses (shown on the same graph) indicate actual crossover points of 860 Hz and 3.1 kHz. The horizontal off axis response is quite remarkable. From 400 Hz to 8 kHz, the measurements indicate very nearly uniform dispersion. In the case of the Neo8 and Neo3, this is understandable as the drivers are the same technology and same width, but I was not expecting the transition from the ND140 to Neo8 to be that uniform. The impedance plot shows the cabinet tuning frequency of 38 Hz. Although this design features two 8 ohm drivers in parallel, the minimum impedance is right at 4 ohms at 130 Hz. This should be a reasonable load for most amplifiers to handle.

    Impressions:

    The current incarnation of the Matrix has been a lot of fun to listen to. The spooky micro detail of the original is still there, but the bass is now also fully present and accounted for. These were demoed at DIY Iowa 2017, so others may chime in with their impressions.

    Showtime Crossover Schematic On axis frequency response
    Keep an open mind, but don't let your brain fall out.

    Argon | Helium | Lithium | LithMTM | Hafnium
    Mercury | Matrix | Shrubbery

  • #2
    Cool! Now THAT'S a narrow cabinet.
    Isn't it about time we started answering rhetorical questions?

    Paul Carmody's DIY Audio Projects
    Twitter: @undefinition1

    Comment

    Working...
    X