Stuff it with pink fluffy?
And make sure no one EVER uses it.
The obvious question is, since it is never used, why not consider removing it? it may come down to a choice between good sound and room esthetics. Choose wisely--you will be judged!
I think, Stuart, when we do the AVS Oscars in February, you will definitely be nominated for "Deepest Into The Rabbit Hole, 2013" category. :)
We've done it. It's not that bad. You need a plumber to cap off the gas line, then just about anyone can take out the stove. You might even be able to sell it. We left the chimney in the roof so we wouldn't have to seal that up, just the ceiling in the room.
OK. I spent today recalibrating with Audyssey Pro and then measuring with REW following 'doubling up' on the insulation in all of my corner traps and adding a new wall/floor corner trap (also doubled up) and two smaller custom-sized traps along the front floor/wall corner, between the speakers. The latter have 4 inches of insulation only due to space considerations.
This is the waterfall before the traps were upgraded - mic position is as close as possible to that used today for the 'after' measurements, as is SPL.
This is the waterfall after adding the additional absorption:
There is some improvement but not where I expected it to be! I was trying to get rid of the problem at about 50-60Hz yet that seems pretty much unchanged. Comments welcome.
Showing the importance of the measuring SPL, here is an 'after' waterfall made from a measurement taken at slightly lower level:
Big 'graphed' difference.
Frequency response was also changed a little but remember the mic was not in exactly the same position as for the earlier measurements, just as close as I could get it. All graphs below unsmoothed.
Bass region before adding additional absorption:
And both overlaid to make comparison easier:
I would welcome comments and observations from the thread - especially with regard to whether anyone thinks the effort and expense of upgrading the treatments was worth it, or not. Thanks.
BTW, how does it sound? Bloody marvellous!!
An interesting and challenging room!
Regarding the big dip at 7K, this could be caused by interaction between your left and right speakers, especially if the mic is not exactly centered between the two. Centering the mic is difficult, and you can get big swings in the higher frequencies. Two suggestions:
1. Measure Left+subs, and then Right+subs. If the dip does not appear in these two graphs, then the dip in the combined graph is probably due to interaction.
2. Place the mic in what you think is the center position, run a measurement of Left+Right+subs, move the mic a little to the right (only about an inch!), repeat the measurement, then move it to the left a bit, re-measure, etc. Observe whether the dip at 7K changes when the mic is moved. This is another indication that the dip is caused by interaction.
The waterfall actually looks good based on the parameters you have entered. To dig a little deeper, do this:
1. Open the RTA tool in REW, configure using the settings shown below, and click the red button to start the measurements. Let it settle down (after averaging 32 measurements), and observe the noise floor in your listening room. Of course, you should do this with as quiet a room as possible. The noise floor should be somewhere between 40dB and 50dB.
2. Configure the Waterfall vertical axis lower limit with the noise floor value you observed in step 1. (You currently have 50dB configured, and this may well be the noise floor in your room.)
3. Configure the Waterfall window to 450ms (vs. 600ms).
4. Generate a new Waterfall and post the results.
It's not as bad as it looks. Unsmoothed trace there remember. When it's 1/6 smoothed it's just a dip of a few dB. And it's pretty narrow. But where it came from IDK.
I wasn't too disappointed as I think there have been minor improvements - but I was hopeful that the problem at 50-60Hz might be improved more - but Sanjay has an informative view on that so maybe I was expecting too much. I don't attach any real significance to subjective listening impressions but FWIW it does sound a little 'tighter' (subjectively). Maybe those small improvements at the bottom end which we can see on the graph make a disproportionately large audible improvement?
The modified traps look just the same as the originals - all I did was stuff a slab of insulation into the air gap. My room is so small that you can't actually see 'into' the traps from the ends from anywhere you would normally stand so there was no need to disguise the insulation at all. The one exception to that is the panel above the screen on the wall/ceiling corner where you can see into the end as you walk into the room - so I just wrapped it in some black Camira I had left over and it has become invisible. It's so freakin' dark in there, even with the lights on, that you can get away with a lot.
Would I go to the trouble and expense over again, knowing what I know now? Possibly. At this level of development of our rooms, we can only expect fairly small incremental gains I think and that is what I have achieved. The cost has been minimal, relatively, (very soon I will be shelling out over $4,000 on a new PJ for example) and at least I know now that I have done all I can in this regard.
Would I recommend anyone else doing the same? Not really - there may be better ways to spend the money, depending on the current gear etc the user has.
Thanks Sanjay. Yeah, traps 2ft thick are, of course, impossible in such a small room. I'd be happy to use such traps if the room could accommodate them, but it is what it is. Given what you say, and what your informative diagram shows, I will never be able to do much about that small 54Hz peak. I can't move the seating far enough forward to make much difference. Your memory is good - the entry door is indeed along that right wall, so that rules out potential solution. The other options are, I think, too much effort for a fairly small gain (and may bring another corresponding issue with them at some other frequency I would guess. Thanks for the helpful post and ideas though. I always value your interventions.
LOL. I fear that it may be too heavy for the floor, let alone the door
Sanjay has a good memory. I have already blocked off and concealed one door into the room (on the left wall/front wall corner) which took some negotiating skills I might add :) The only other entrance is the door on the right wall so I can't block that off or I would have to enter and exit the room via the window. Oh - I can't do that as the window is blocked with blackout blind and also has two GIK 244 panels on the windowsill, behind the curtains, which are permanently drawn of course. :) Most Americans would laugh their socks off at the size of some of our European rooms!
I see nobody commented about the difference between these two waterfalls, with about 3-5 db lower SPL level in the second "after" waterfall colored in green vs. the top one in purple. The difference in the charts demonstrates how a small difference in measurement may lead to radically different interpretations about the implications for your room.
Yes indeed - I made the second 'after' chart deliberately to see the difference made by using a lower SPL for the measurements. It makes a big 'cosmetic' difference doesn't it?
Yes - I swear I can hear a difference, but I know too much about expectation bias and sighted tests to place much store in that. I am happy anyway as the result is fairly good, even after applying the old diminishing returns law.
Yes, I would normally measure to about 95dB peak and use a noise floor of 45dB. That way I hope I am not cosmeticising the graph and deceiving myself. Here is the same graph with the noise floor limit set to 50dB. Looks even better of course :)
Funnily enough, I was pondering that very question while I was measuring yesterday. For example, here is the waterfall with the slightly lower test SPL, 50dB noise floor and now with a 300ms time slice applied... looks not too bad (other than my known and probably unresolvable issue at 50Hz):
I didn't ponder this for too long yesterday as I normally listen at -6dB from reference, so I am peaking at 106dB (bass) and 99dB average if all other things are equal. But those peaks only last for a few ms of course, so maybe we do attach too much importance to them. The real world is not the graphs world.
Yes, I agree. The first chart in this post (45dB noise floor, 95dB peak) is the one I would normally use. Of course, even those of who ARE watching action movies at very high SPLs, the issue is a bit 'meh' - who cares if the explosions ring a bit? Do we really hear it that way? Or is it just more 'explosion'. I do want any bass in the score to be accurate and as tight as possible - but it will be as there isn't a lot of music below 40Hz anyway. And I like the mid and upper bass to be nice and tight... but for splosions, does it matter all that much? I'm not saying ignorance is bliss either - just putting a 'real world' context on it like you are. And how many movies in any case have much real content below 20Hz anyway? Some for sure - but by no means all, or even many.