Home Theater Forum and Systems banner

41 - 60 of 68 Posts

·
Registered
Joined
·
2,261 Posts
Discussion Starter #41
UPDATE:
This last weekend was a mixed bag of tricks in terms of progress. On one hand I was able to rearrange the room to take acoustic measurements for two more layouts. On the other hand, I encountered a major setback when discovering a dead left-channel tweeter. Something went wrong between the first listening trials and measurements for Layout-1. Long story short: a new tweeter is on its way. That's the bad news. The good news is that the blown tweeter explains the wide response dip around 3.5kHz (the rising response from there to 20kHz is due to the rear tweeter).

The other room arrangements I tried basically experiment with subwoofer location. Layout-2 swaps album crate and subwoofer locations. It has three stacks of crates along the back wall, and the sub along the left sidewall. Layout-3 puts the sub between the last stack of album crates and the door, leaving just enough room for a person to squeeze in. Listening trials for these layouts didn't last long, partly because of the blown tweeter, and partly because of my resultant foul mood. Despite the setback, I listened for bass quality and measured low FQ. The graphs showed a steep 20dBSPL drop from 40Hz down through 15Hz, and many peaks/dips from 40 through 300Hz - not at all like the fairly flat response observed for Layout-1. Different locations along the sidewall did not help. Listening to my bass test tracks told me all I needed to know:
  • Bass was generally boomy
  • Subwoofer could be frequently localized
  • Punch factor was gone
Realizing that lowered treble can be perceived as an increase in bass and vice-versa, I'll conduct more listening tests after repairing the tweet. But I doubt an improved frame of mind is going to change the overall bass quality I heard in music and movies from Layouts-2 and 3. I'd like to try one more sub location right next to the LP (Layout-4) before declaring Layout-1 as the SQ winner. I can publish some graphs if anyone is interested, just ask. :D
 

·
Registered
Joined
·
2,261 Posts
Discussion Starter #44
Glad you got to experiment with the sub though. I hate guessing with stuff.
Yeah, same here. And speaking of guessing, I'm still a novice when it comes to many things HT, subwoofer setup included. I thought that moving a sub a few inches can make a big difference, but I'm finding that's not the case (at least in my room so far). Guess I could have saved myself a lot of REW measurement heartache by moving the sub a foot at a time instead of inches. Do you know if that should make a difference? I'd like to know because I'll need to start my subwoofer location tests all over again.

But why, Lou? Well, because I made another mistake, but this one brings good news! My curiosity got the better of me last evening... how could bass plummet from satisfying into boomy, localized slop? What changed? Then I remembered I engaged the LOUDNESS function for a late-night movie, thinking it would only be active for that "session." Boy was I wrong! The LOUDNESS setting stayed active through the next power-up sequence. I was expecting it to act like the L/C/R/Sub trims, which only retain their setting for the active session. That'll learn me to expect logic from an AVR! So just take a wild guess what happened during my next listening session when the volume was cranked with LOUDNESS engaged. Yup, an overabundance of low-quality bass which I chalked-up to subwoofer positioning instead of an AVR setting. And to make matters worse, I realized my rear "wall" was now a matrix of filled album crates. My measurements and listening tests were being conducted from a "secret" LP much closer to the back wall than was thought or planned. It's no wonder turning down the subwoofer trim didn't help much. The good news is the drop in bass SQ now has a good explanation.
 

·
Registered
Joined
·
3,991 Posts
Well Lou. IME moving a subwoofer by inches instead of feet yields very minimal changes. I'm sure that's dictated by room size. In my gymnasium, even one foot shows no changes in measurements. But in my bedroom smaller changes are evident.
So glad you figured out the boom and bloom! I HATE when settings don't go back after powering down. I've been burned by that too. As to the LPs.(sufferin mind), I'm kinda guessing, but imo the density created by stacking them tightly in crates, would crest a barrier. Probably even more so than drywall and wooden studs.
 

·
Registered
Joined
·
3,991 Posts
Lou, Lou, Lou...
If we started making fun of you, you might stop sharing!!! Lol!!!
Most of us aren't professionals either(of course some are), and there is much to be learned from your trials and tribulations. Doin good!
 

·
Registered
Joined
·
1,784 Posts
Not making fun of you ?? Why nonsense my friend, I would surely make fun if I was smarter and could understand your missive's. I am left behind in the dust here. :scratch:
 

·
Registered
Joined
·
2,261 Posts
Discussion Starter #49
Well, at least I can laugh at myself! You guys are too kind; thanks for the encouragement. Now I'm starting to understand why you called me brave! What was I thinking? But I'm not ready to crawl away in shame yet!!

This thread isn't turning out as planned. The forum's big fish aren't taking my measurement bait - maybe because the thread's title needs focus, or maybe because it seems off-topic to REW measurements. In any case, I'm moving it to a less-technical forum and will continue to post measurement questions individually.
 

·
Registered
Joined
·
1,784 Posts
I am sorry Lou but I cannot do the measurement thingie just yet. Next time we get together you can show me or....maybe I could just read the manual :laugh2:

I dont learn as well by reading something as by doing it...so you have left me done and dusted. :crying2:
 

·
Registered
Joined
·
2,261 Posts
Discussion Starter #51
The tweeter's failure mode leaves me with a nagging sensation. Why hadn't I noticed it sooner? Of course, hindsight's always 20-20! It's clearly evident in the SPL graph, but on my behalf the tweeter hasn't totally given up the ghost. It still functions at the lower end of its crossed frequency range! The dip is not as noticeable to these ears as the graph would have me believe. Again to my credit, I think it took a while to detect because of the masking effect presented by the right and center channels. I've never heard of such a failure mode, though. Can a tweeter fail "halfway" so it's only reproducing part of it's normal range, or would that be a crossover problem?
Layout 1 FR - LR Only.jpg
 

·
Registered
Joined
·
2,261 Posts
Discussion Starter #52
I am sorry Lou but I cannot do the measurement thingie just yet. Next time we get together you can show me or....maybe I could just read the manual :laugh2:
I dont learn as well by reading something as by doing it...so you have left me done and dusted. :crying2:
You're not alone. Seeing is easier than doing for me as well. Once I get the hang of it, I won't mind sharing; so no worries, we can take a look at your setup during our next audio get-together!
 

·
Registered
Joined
·
1,784 Posts
The tweeter's failure mode leaves me with a nagging sensation. Why hadn't I noticed it sooner? Of course, hindsight's always 20-20! It's clearly evident in the SPL graph, but on my behalf the tweeter hasn't totally given up the ghost. It still functions at the lower end of its crossed frequency range! The dip is not as noticeable to these ears as the graph would have me believe. Again to my credit, I think it took a while to detect because of the masking effect presented by the right and center channels. I've never heard of such a failure mode, though. Can a tweeter fail "halfway" so it's only reproducing part of it's normal range, or would that be a crossover problem?
View attachment 138634
Good Morning Lou

I am not sure but it would be my thought that either the tweeter works or it does not. I am not sure I have ever encountered a driver like your tweeter working over a selected frequency and not the rest of its inherent range.

I wonder if maybe there is an actual crossover issue in the one speaker ? Let me rephrase that statement, It would appear that there may be a crossover issue in the one speaker.
 

·
Registered
Joined
·
2,261 Posts
Discussion Starter #54
I initially considered drivers to be digital in nature - that is, they either work or they do not. However, further investigation reveals that some engineers believe in partial failure modes which leave the driver damaged, yet operational. An example of a failure mode similar to my own appears in this Speaker Failure Analysis article, where the author states:

"For a host of reasons, glue attachment of the voice coil to the cone or spider may fail. The immediate result of even a partial failure is the voice coil is no longer precisely centred. An off-centre voice coil will scrape against the pole pieces and this soon damages the wire it is wound from, possibly creating shorted turns in the process. This highly audible defect is called "poling" by repairers. Broken or cracked attachments vibrate severely under normal drive forces, so this is also highly audible as buzzing noise or distortion. Repair is sometimes possible but re-coning or replacement is often needed."

While I have heard HF distortion, it's difficult to tell whether it originates from the tweeter, midrange, or source material. But that still doesn't explain the main failure mode of partial output. I did find related posts on the Electrical Engineering Stack Exchange, which support my recent experience:

"...However, one can imagine a scenario where something else in the speaker is damaged, such as the surround (the rubber coil that allows the cone to vibrate freely) or the spider (cloth "guide" inside the speaker, behind the cone) or leads leading into the cone are damaged, but the coil still operates. Any of these issues have the ability to reduce the volume level of the speaker."

and

Another part of the speaker that can fail is the joint between the coil and the cone. If the force on the coil exceeds the strength of that joint, it's possible that it may fail partially or completely. If this happens, motion of the coil will not be properly conveyed to the cone, causing the speaker to sound rather weakly. Additionally, the speaker will likely be more effective at producing sound when driven in the direction that would pull the coil against the one, than when driven in the opposite direction. Thus, the sound will not only be quieter than it should, but distorted as well.

I know there are smarter and more experienced enthusiasts than I who may present evidence and arguments against what I have found so far. But in the end, I still have a partially functioning tweeter, whose failure mode was verified by swapping drivers in the left and right mains - the problem followed the driver! Installing a totally new driver in place of the malfunctioning one solved the problem. This should be confirmed with my next round of measurements.
 

·
Registered
Joined
·
2,261 Posts
Discussion Starter #56 (Edited)
Please feel free to add your comments and/or suggestions! These charts show the improvement realized first by adding 2 bass traps (2T), and then by adding 3 more (5T). As more traps are added, frequency response starts flattening out and resonant modes start decaying more rapidly.

Frequency Response Improvement (Left Channel)
Maybe quotes should go around the word improvement. Adding traps sure doesn't make the FR plot look pretty. You'll need to turn to the waterfall for that. A quick glance at the Sub-Only plot shows that the sub dominates the combined responses (L+R+Sub and Center+Sub).
Layout 4 FR (0T2T5T) - L Only.jpg

Frequency Response Improvement (Right Channel)
Same here! Maybe quotes should go around the word improvement. Adding traps sure doesn't make the FR plot look pretty. You'll need to turn to the waterfall for that. A quick glance at the Sub-Only plot shows that the sub dominates the combined responses (L+R+Sub and Center+Sub).
Layout 4 FR (0T2T5T) - R Only.jpg

Frequency Response Improvement (Center Channel)
My novice eye has yet to detect much improvement here as well. The traps' effectiveness down low is extremely limited to nonexistent, as is expected.
Layout 4 FR (0T2T5T) - C Only.jpg

Frequency Response (Subwoofer Channel)
As expected, there's only slight improvement where FR is leveled from about 75 to 150Hz. Larger and/or more traps would be needed to make a significant dent above 40.
Layout 4 FR (0T2T5T) - Sub Only.jpg

Bass Resonance Improvement (Left+Right+Sub Channel)
LEFT = 0 Traps & 2 Traps / RIGHT = 0 Traps & 5 Traps

These waterfall comparisons are best viewed "live" in REW, but I tried to make them as revealing as possible. The baseline measurement with 0 traps is always the same color, and is shown transparent. Examining the first slice of the 5T plot reveals that FR is smoothed from around 65Hz to 120Hz, and to a lesser extent from around 120Hz to 300Hz. Noticing where the distance increases between succeeding slices tells us if and where ringing has been improved. Ringing has been totally subdued by trapping at around 300/210/180Hz and partially subdued at 205/150/130/95/75Hz. Severe ringing still exists at 45/22/19Hz, which are too low for traps to be effective.
Layout 4 WF (0T2T) - LR S12.jpg Layout 4 WF (0T5T) - LR S12.jpg

Bass Resonance Improvement (Center+Sub Channel)
LEFT = 0 Traps & 2 Traps / RIGHT = 0 Traps & 5 Traps

I'm unsure if this speaker combo even needs scrutiny. I only included it because it was a recommended measurement in the REW 101 Guide. As before, these waterfall comparisons are best viewed "live" in REW. Examining the first slice of the 5T plot reveals that FR is smoothed from around 70Hz to 170Hz. Noticing where the distance increases between succeeding slices tells us if and where ringing has been improved. Ringing has been totally subdued by trapping at around 200Hz and greatly improved at 150/125/95/75Hz. Severe ringing still exists at 45/22/19Hz, which are too low for traps to be effective.
Layout 4 WF (0T2T) - C S12.jpg Layout 4 WF (0T5T) - C S12.jpg

Decay Improvement (Left+Right+Sub Phase=0)
LEFT = 0 Traps / RIGHT = 5 Traps

On the left we see that ringing at 45Hz does not decay 20dB within the recommended 160msec target. On the right we see that 5 traps brought the decay under control even though ringing still exists! This is actually proof of the traps effectiveness in action. To quote The ABC's of ASC's Tube Traps: "It’s worth keeping in mind that ASC’s tube traps aren’t bass absorbers per se, they are resonant absorbers. They won’t remove the bass energy from your room, rather they will tame the resonances caused by this energy. Bass peaks will still be strong at the room’s resonant frequency, but the absorption of this resonance will reduce their severity."
Layout 4 DECAY (0T) - LR S12.jpg Layout 4 DECAY (5T) - LR S12P0.jpg

Decay Improvement (Left+Right+Sub Phase=90)
Dialing in the sub's phase control for proper support through the 80Hz xover region shows that decay times are greatly improved and well within target.
Layout 4 DECAY (5T) - LR S12P90.jpg
 

·
Registered
Joined
·
2,261 Posts
Discussion Starter #58
Results of the room/system calibration and center channel upgrade are impressive so far:
  • line-of-sight to surrounds established for first time
  • semi-nearfield LP eliminates many damaging reflections
  • bass has regained it's punch
  • speech intelligibility has been greatly improved
Yet I'm left with only a lukewarm feeling. The system could be taken up another notch, but I'm not sure how to proceed. Maybe it can't be done with my current gear (is a miniDSP in my future)? I'm hoping to learn enough from my REW threads to find out!
 

·
Registered
Joined
·
2,261 Posts
Discussion Starter #60
And just to put your suspicions to rest, no it's not business-as-usual. I plan on taking your advice rather than asking for it, but then doing what I want anyway. The current 5-trap layout with sub next to the LP is a temporary solution until I develop something better. So far, it's my favorite.

Room Layout #1 (original before this renovation):
Well-integrated bass with reasonably flat FR and excellent support through xover region. Fair SS&I. But too many traps required, so fairly lifeless.

Layout 1.PNG


Room Layout #2:
Very poor LF response. No amount of good symmetry/aesthetics makes up for sound this bad!!

Layout 3.PNG


Room Layout #3:
Good LF response with engaging tactile feedback. Very good SS&I. Least aesthetic of options, but who cares!

Layout 2.PNG


Room Layout #4:
Fair LF response with some extreme LF feedback (expected more sitting right next to sub). Very good SS&I. Good symmetry/aesthetics.

Layout 4.PNG
 
41 - 60 of 68 Posts
Top