Home Theater Forum and Systems banner

1 - 4 of 4 Posts

·
Registered
Joined
·
39 Posts
Discussion Starter #1 (Edited)
Hi All,

Up front apology for the long post with some busy graphs. :bigsmile: (REW is GREAT!)

I have finished up a DIY Transmission Line subwoofer and planning to integrate into the rest of the system as best as I can. System details below.

Speakers - Definitive Technology - Front L/R BP2002TL, Rear side L/R BP2X, Rear Back CLR2500, Subwoofer - DIY - Transmission Line @ 30Hz, 12" Eminence, O Audio 500W rms plate amp, AVR - Denon - AVR-3801, Optical - OPPO - BDP-95, HDTV - Panasonic VT50 65"

What could be done for 60Hz dip? Since this is a sub evaluation should I be concerned much about anything above 80Hz? No averaging or smoothing is applied to any of the graphs.

Each front tower has its own sub while the DIY is placed far behind the listening position. Room is 40ft x 17.58ft x 9ft (DxWxH)
First graph is all three subs measured at six seat positions across the couch. I have limited EQ ability but might be able to bring the peak at 44Hz down some. The dip just below 60Hz is worst at seating position 1 closest to the left wall. Multiple subs helps some. The dip just above 80Hz show up shows up multiple positions. Typically all the subs are crossed over above 80Hz. Might have to play with front tower position and see if this upper mid based dip can be fixed.
all_pos_all_sub.jpg

Here is the Transmission Line itself. LPF is disabled for first sweeps. Based on internal dimensions is should not be used for frequency much above 80Hz. So can't tell is the 80Hz dip is room or enclosure related in these sweeps.
all_pos_tl.jpg

Here is the Left Front Tower. SPL varies a lot based on seating position.
all_pos_dtl.jpg

Here is the Right Front Tower. 30Hz, 60Hz and 85Hz have dips. Thankfully the 30Hz dip goes away when all subs are playing.
all_pos_dtr.jpg

Huge Thank You for any comments/suggestions.
 

·
Banned
Joined
·
4,838 Posts
Your room modes are always a factor. Plus sub/spkr placement relative to walls and ceiling (hard to do much with sub/spkr-to-ceiling spacing:bigsmile:).

For instance, having sub placed so front of sub is out from the wall about 4.7 ft will give a cancelling wavefront reflected from the wall and delayed about 8.3 mS, giving a cancellation and notch at 60 Hz. In summary, 60 Hz notch could result from a delayed path length of 9.4 ft (8.3 mS) - that is a delayed path length relative to the.direct sub-to-LP path length.

By the same logic, the 82 Hz notch could result from a reflection path with 6.1 mS delay, or 6.9 ft.

That is one way to look at things, although room modes can be far more mysterious than that simple view. One diagnostic approach is to move the sub a foot farther from the wall and see what changes. Or lift it midway between floor and ceiling :bigsmile: (I have seen it done). Or just turn it 180 degrees.
 

·
Registered
Joined
·
39 Posts
Discussion Starter #3
Your room modes are always a factor. Plus sub/spkr placement relative to walls and ceiling (hard to do much with sub/spkr-to-ceiling spacing:bigsmile:).

For instance, having sub placed so front of sub is out from the wall about 4.7 ft will give a cancelling wavefront reflected from the wall and delayed about 8.3 mS, giving a cancellation and notch at 60 Hz. In summary, 60 Hz notch could result from a delayed path length of 9.4 ft (8.3 mS) - that is a delayed path length relative to the.direct sub-to-LP path length.

By the same logic, the 82 Hz notch could result from a reflection path with 6.1 mS delay, or 6.9 ft.

That is one way to look at things, although room modes can be far more mysterious than that simple view. One diagnostic approach is to move the sub a foot farther from the wall and see what changes. Or lift it midway between floor and ceiling :bigsmile: (I have seen it done). Or just turn it 180 degrees.
Thanks for the input. You may have hit the nail right on the head. :T

I re-read a section of a Floyd Toole book that discussed wall mounted speakers. Basically the distance from speaker to wall that correspond to a half-wavelength results in a notch in the frequency response.. Did the math and confirmed your calculations.

The very interesting part is that the distance from ceiling to the bottom (or just beyond) my mid range speakers corresponds to 82-85Hz. Same deal for ceiling to sub (port). 54-58Hz corresponds to the distance from ceiling to floor (port exits at floor level) or a little beyond.

So these two modes appear at slightly lower frequency than the math would suggest. I think this might be normal too for residential walls that have some flex to them.

Also the room height and width are almost multiples of each other. This could be compounding the effect.

Hmmm - Might be some DIY speaker stands in my future. Moving the speakers up will shorten the distance but only push the modes to higher frequencies. Maybe the depth of the notch will also be reduced (lower Q) if the compounding is also less?
 

·
Registered
Joined
·
39 Posts
Discussion Starter #4
Well I was able to find out how to resolve the null around 85Hz.

I just simply had to turn on the X-Over on the Transmission line plate amp and "bingo" no more dip.

Attached shows a reference sweep from January (light purple) plus two sweeps with the setup in application. These two sweeps are red/maroon color. I.e. X-Over ON. Followed that up by disabling the X-Over yet again to confirm my suspicion. This sweep is in blue. We call this an A-B-A test in my industry.
I just doesn't get any easier than this.
:T
 

Attachments

1 - 4 of 4 Posts
Top