Speaker Repositioning Project

boxerfan88

Well-known member
After procrastinating for a couple of months, found some time over the long weekend to start the speaker repositioning project.
  • remove the center console
  • move the subwoofer to the middle
  • push the mains L+R closer to front wall
  • pull apart the mains L+R to get closer to 60° angle

Speaker Placement 2024 - v1.png


Results:
  • Pros: soundstage becomes wider, bass is more even, images spread out a little bit more across the soundstage, tonal balance similar to before.
  • Cons: the room is a little more lively / has a little bit more flutter echo because the 2 corner bass traps & console were removed.
 
Last edited:

boxerfan88

Well-known member
From the measurement sweeps, we can see shifts in the peaks/dips frequencies due to repositioning of the speakers.

The mess in the 100Hz-200Hz region is still there but they have shifted.

Here's the "NoEQ" chart:

NoEQ LR.png

And the "with EQ" chart:

EQ34e LR.png

.
 
Last edited:

boxerfan88

Well-known member
Excess Group Delays shows minimum phase behavior in most areas except for the messy 100-200Hz room transition zone.


There are four frequency zones that have an impact on the way sound moves through a room. In a typical residential space, the second zone, also known as the Schroeder frequency (which ranges from 100 to 200 Hz) is the transitional or crossover zone where room resonances dominate until wavelengths adjust to the size of the room.

Energy below the Schroeder frequency is wave energy (or low frequency), which is like static air pressure and is handled by a resonator. Energy above this zone is ray energy (or higher frequency), which bounces around the room like light and is handled by a reflector. Note that there is also a third zone that serves as an additional transition between the Schroeder frequency and the fourth zone.




L+R GD.png




.
 
Last edited:

boxerfan88

Well-known member
IMG_1201.jpeg



Trying out the ikea bag holder as cable lifter to lift most cables off the floor, so as to facilitate cleaning in the future.
 

boxerfan88

Well-known member
Since this project resulted in the temporary removal of the bass trap, I had the opportunity to measure the audio amplitude response without & with the bass traps.

The charts below show that bass traps definitely helps in the bass region (Schroeder room transition region).
  • NoTrap = no bass trap in the room.
  • wTrap = 2pcs MOAB bass trap inside the room, one at front left corner, one at front right corner.
  • the bass traps reduces the crazy deep cancellation dips quite a bit at the Schroeder region & crossover region.

NoTrap wTrap LR compare.png


I also decided to rotate the subwoofer back to front-back orientation. I didn’t like the idea of subwoofer side firing directly into the AVR. 😅

.
 
Last edited:

boxerfan88

Well-known member
After 2hrs of listening to all sorts of tracks ... clearly getting close to the ideal 60° angle / equilateral triangle really helps open up the sound stage.

Also, the "depth" is a little more apparent with this speaker positioning -- I was kind of worried that the "depth" would collapse if I moved the speakers almost right up to the front wall. I'm happy that it turned out very well.

.
 

boxerfan88

Well-known member
Was poking around the charts comparing the before repositioning vs. after repositioning, looking for obvious differences.
I noticed the impulse graph was dramatically different.
The “before” chart, after the initial impulse, the signal still jump around quite a bit (high variation).
The “after” chart, the initial impulse looks tighter & sharper, the variation is a lot lesser (orange & blue).

Imp LR.png
 
Last edited:

boxerfan88

Well-known member
New positioning spectrogram chart ... reasonably quick decay times of 300ms from 200Hz onward, looks pretty even as well. 300ms I think is sufficiently lively for hifi.

Allows the musical content to shine through with minimal room resonance/flutter echo overhang.

Still have some room resonances in the lower region <200Hz, some hanging around for almost 700ms.
Thus far, not really affecting the music enjoyment ...
Maybe one day soon I can test at higher SPL to see if those low frequency resonances is a problem or not.

Spectro LR.png


Decay times that either too long, too short or are notably different to the decay times in other frequency bands all cause audible sound quality degradations:
Too long - overly long decay times, even when obvious echoes cannot be detected, results in a loss of low level detail in much the same way as a room with a high noise floor. This is because low level sounds within the music are obscured by the slow decay of other louder sounds. This phenomenon is also known as masking and can be employed for good in noise nuisance reduction systems. Slow decay times consequently reduce the ability of a system to reproduce critical acoustic cues such as the decay character of a recording venue or the precise location of a musician within the soundstage. A room with long decay times also tends to sound harsh and brittle and can be an unpleasant place to listen resulting in rapid fatigue. Overly long decay times are considered those over 0.5s.
Too short - short decay times can result in a lack of spaciousness and envelopment. Overly short decay times are considered those under 0.2s. Like most things in acoustics, however, things are not simple. Two rooms with identical single figure reverberation times can sound quite different depending on where the absorption is located. For example our showroom home theater has very little mid/high frequency absorption on the lateral surfaces but a lot on the ceiling and floor. That room will sound different to a room with a lot of absorption on the side walls and not much on the ceiling or floor.


 
Last edited:

boxerfan88

Well-known member
Taking a look at RT60 (Topt) charts.

Before chart - very messy and both not aligned in the 60Hz-200Hz region.

RT60 Topt Before LR.png


After chart - both are better aligned in the 60Hz-200Hz region.

RT60 Topt After LR.png

.
 

boxerfan88

Well-known member
With the new speaker position, the soundstage is much larger and more enjoyable. More space for the images. Depth seems to be slightly better too. (seriously if you have not setup in an equilateral triangle, try your best to get as close to it as possible, it is worth the effort.)

After a week of listening, some bass heavy tracks can be a tad boomy. More EQ fine tweaking this weekend, this time targeting the room ringing...

Result not bad lah.

Spec LR.png

EQ34 LR (var).png
 
Last edited:

boxerfan88

Well-known member
Interesting observation on the Impulse graph.

There is a big reflection around 5.1ms after the main impulse, which translates to about 1.75m. This means the reflection path length is 2.6m+1.75m. I wonder which surface is it 🤔


Impulse LR.png
 
Last edited:

boxerfan88

Well-known member
For the fun of it, measured at 5 locations at the MLP seat (front, center, rear, left, right)

Quite interesting to see the variations in the charts.

In real life, when moving the head around the MLP seat, for example lean forward, or lean sideways, the tonality does shift slightly. However the differences heard is much lesser than what the chart shows, which means our brain is pretty good at tuning out some of the room characteristics.


MLP-LR.png

.
 
Last edited:

boxerfan88

Well-known member
I was reading up on SBIR and about the +6dB front wall gain.

I revisited the "before EQ" REW measurement ... and was pleasantly surprised by the uncanny match.

"Theory on the left." -|- "REW measurement on the right."


sbir-3.png


.
 
Last edited:

boxerfan88

Well-known member

boxerfan88

Well-known member

There’s nothing inherently wrong with large EQ boosts if the result sounds good — but if they’re more than about 6dB, it’s probably a sign that there’s a better cure somewhere else!
 
Top