Newform Research, SB, and Dayton Towers

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I've just finished the construction of these "somewhat" over the top speakers, save for some Minidsp tweaking. Just want to listen to them flatish before adjusting.

Woofers are Dayton Audio RS225-4
Mids are SB Acoustics SB15MFC30-8
Tweeters are a stack of 2 - 30" and 1 - 15" Newform Research Oval series.

I have some build photos coming. Today I have some more listening to do.
 
The Pics
 

Attachments

  • 20180806_193007.jpg
    20180806_193007.jpg
    137.8 KB · Views: 376
  • 20180806_193038.jpg
    20180806_193038.jpg
    103.1 KB · Views: 376
  • 20180806_193051.jpg
    20180806_193051.jpg
    97.3 KB · Views: 372
  • 20180806_193104.jpg
    20180806_193104.jpg
    138.1 KB · Views: 364
Here are my preliminary measurements, but still tweaking the crossover points, slopes, etc...
The graph shows 1/12 smoothed at listening position.

Crossover points are:

Mid to high = 1150hz with 48db slopes.
Mid to low = 125 hz with 12db slopes.
Low to sub = 45 hz
Mids to highs are so low to deal with lobing.

I built them so tall to have equal reflections with a standard 8' ceiling.
 

Attachments

  • Speaker Measurements.png
    Speaker Measurements.png
    121.3 KB · Views: 253
Chances are you have some early front and side wall reflections corrupting your frequency response. First look at the impulse response and see the time of arrival of reflections. If the reflections are early and strong, then some absorption panels might be in order. Then press the IR windows button at the top of the REW screen and configure frequency dependent windows. That should make it look a lot smoother and show you what the direct response is absent the reflections.
 
If green and red graphs are representing left and right speaker, than something strange is happening with one of the speakers at 4.5 and 9 kHz (red curve), because green and red curves should be very similar above 1 kHz.
Mid to high crossover is not optimal - there is a dip just above 1 kHz. Try tweeter delay or some phase shift with MiniDSP.
Very impressive build!
 
I spent some more time on them today, this time with a few differences.

- Used some room treatments to tame reflections.
- Measured at 5'6" from the speakers as they are so tall I wanted to avoid the woofers being too far off axis.
- Measured each driver separately, PEQ'ed major issues. Attached graphs after this adjustment.
- Found some definite variation in the Newform tweeters left to right. Photo attached.
- Mids too had some variation, but not as extreme.

- Last graph is at listening position, L+R. 8" delay set on the left speaker due to slightly off-center room. Subwoofers included set quite high (90hz) to deal with a 80hz mode I without them.

- Used all BW 24db crossovers.
- Had some big dips in the mids. Oddly I inverted on side of the mids and found that cleaned the dips up. I used a battery to check to see if I had accidentally mis-wired one side but they were good. I'd love to hear thoughts on this.
- When I set up to listen to them I switched back and forth with the inverted channel. Found that the sound seemed fuller with both in phase, but vocals were more pronounced in the right speaker. Played with delays, never helped. Inverted one side, vocals moved to center. Grrr.

- Post some listening I inserted a wide 3db dip at 2500 hz and a 2 db low shelf increase starting at 170 (for my preference).
 

Attachments

  • Highs.jpg
    Highs.jpg
    65.4 KB · Views: 55
  • Mids.jpg
    Mids.jpg
    68.6 KB · Views: 43
  • New Measurements.jpg
    New Measurements.jpg
    67 KB · Views: 37
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.