I'm going to do a DIY absorber panel, but have some questions... (1 Viewer)

[Jones_Rush]

I want to go according to the plans here:
http://pat.home.mchsi.com/SAP.html

This is according to Jon M. Risch's plans, right ?.

If I will go with Miraflex instead of ordinary fiberglass (health issues). Should I need to expect weird consequences for this choice ?.

According to this page:
http://www.owenscorning.com/around/i...s/miraflex.asp

The thickness of Miraflex is 8 3/4". But the question is, will Miraflex absorb as good as regular Fiberglass ?. If they absorb more or less the same, then how come people are still using regular fiber glass ?, is there a really big price difference between the two ?.

One last question, about the absorption coefficient of this DIY panel. Has anyone run any measures to see that the absorption coefficient is more or less constant throughout the entire frequency band ?. As long as it's 0.9/0.8 and above (from the starting frequency) then it should be good. But if it has variations, like from 1.2 at 900hz, to 0.5 at 4000hz, then it can't be good.

Whith commercial pannels I get a frequency table which always shows me a more or less constant absorption coefficient, so I believe this is pretty important in order to get good subjective results...

 

[ThomasW]

Rush

This is according to Jon M. Risch's plans, right ?.

Nope....

Pat's panels are copies he built after seeing mine. And mine were built decades before Jon Risch posted his. They are modeled after a designed created by Jon Marsh, he's a twice published member of the AES.

MiraFlex is fiberglass, so it absorbs like fiberglass. Note that the plastic cover must be removed or the high frequencies won't be absorbed.

Most people have used regular fiberglass because MiraFlex has only been on the market a relatively short time, it's slightly more expensive and slightly more difficult to find

Never not fixated enough to bother about calculating the absorption coeff. I trust my ears with a little assistance from ETF 5.0 and TrueRTA to balance the room acoustics

 

[Jones_Rush]

Never not fixated enough to bother about calculating the absorption coeff. I trust my ears with a little assistance from ETF 5.0 and TrueRTA to balance the room acoustics

Basically, if I'll run a 150hz, 1khz and 15khz sine waves signals (not at the same time) from one side of the DIY panel, and my SPL meter, from the other side of the panel (creates a 90 degree angle with the speaker and panel), will read the same SPL measure, for all three sine waves, then it must mean that the panel has a constant attenuation coefficient through all the frequencies in between the tested frequencies, right ?. Am I missing something ?. If I'm not missing anything, do you think it should be reasonable to expect such performance from the panel, assuming it has been built according to instructions, and was placed far enough from the wall ?.


And another question, I've seen some DIY panels where the wooden frame used was only 3.5" deep, yet the builder tuck 7" of fiberglass into it (and this wasn't the only thing he tucked in). Is this the normal way of getting a 7" fiberglass absorber ?, or, will he just get an improved 3.5" absorber ?. I've seen many DIY'ers recommend the usage of 6 and 8" thick fiberglass. Is this recommendation assumes that you are going to create a 6 or 8" deep frame too ?, if you'll just create a very shallow frame, and try to tuck everything inside using force, will you get the same result as a deeper frame with the same amount of fiberglass ?. To sum this issue: for 6 or 8" thick fiberglass, what is the recommended deepness of the frame ?.

One last question, if instead of a pegboard, I'll use a regular wood without holes at all, so the back of the panel is 100% sealed, will this have any effect on the panel's performance ?. It seems logical that if you say that I need to put the panel away from the wall, it has something to do with not having a completely sealed back, but on the other hand, shouldn't bass below 200hz should easily go through a thin wooden board, even if it has no holes at all ?.

 

[ThomasW]

To do this the right way start by getting a copy of ETF 5.0

Then follow the instructions on their Demo Room Page

Then decide if, where, and how, you should use any sound absorbing panels. And what type of panels are needed.

BTW if you actually need real bass traps look HERE and HERE for info. If you want to make bass traps with fiberglass they'll need to be 12-24" thick.

My DIY panels use 2"X4"s with 6" thick fiberglass batts.

You can use anything you like for the back of the panels.

 

[Jones_Rush]

Thanks for the answers Thomas.

To do this the right way start by getting a copy of ETF 5.0

Then follow the instructions on their Demo Room Page

Then decide if, where, and how, you should use any sound absorbing panels. And what type of panels are needed.

Thomas, I think I'm way ahead of you already .
Since I've started using a room correction device (which made a BIG difference, especially down low), about a month ago, I did tons of ETF measurements of before and after. I even added blankets and pillows at some strategic places in the room, like first point reflections and on the closet behind my head, and could appreciate the difference (I also saw the difference in the ETF waterfall graph, and the impulse response), but I can't say the difference was day and night, it was noticeable (for example, after I covered the closet which is only 2 feet behind my head, with a heavy blanket, I could immediately discern how the imaging bacame more focused at the front of the room. You can say that the 3D effect was reduced, but it was not a natural 3D effect in the first place, since when I go to an acoustic concert, I also hear the band very focused in front of me), but anyway, I wouldn't say the differences took my sound quality to a whole new level.

I could appreciate the difference in imaging, but I'm more after a more accurate and less offensive timbre of the sound. The room is small 12X9x8, wall to wall carpet, and 3 walls out of the four are pretty much bare, while on the right wall there are 3 bookshelves with books. I feel that when I hear the sound, it is like some of the frequencies are a bit colored, or overemphasized to a point, which add kind of a "whistling" effect to some of them, and this makes my ears tired after listening for more than 20-30 minutes at 80 dB SPL (A weighting). I don't think that it is the speakers, since I have pretty accurate speakers (nOrh 9.0s), plus, with the room correction device, the speakers can't produce peaks at all, so it's definitely can't be the direct sound's fault (unless the measurement microphone is not good, but I've heard the ECM8000 should be pretty good, and besides, I hear this "whistling" effect to a greter extent without the room correction).

About the "clapping" test, right after the clapping, I hear a faint but noticeable metalic, high pitched echo, on both ears. I have no clue how it in practice affects the sound of the music, but anyay, it's definitely there with the clapping.

Do you think that if I'll post some pics of my room, plus detailed sketches and some ETF results, I'll be able to get some better help at what I should try to do ?. It's just that after reading so much about room acoustics for the last several weeks, I simply have no clue what to try first, if at all, apart from first reflections.

For example, what should I do with the front wall behind the speakers ?, some people say go LEDE all the way, other say that LEDE is craze of the past, and can only work well for control rooms, so I should just try some diffusion (which other say can't work for small rooms like mine), and other say I should just leave the front wall as it is, since hands down, it's the best compromise possible.

I keep reading contradictions like these for every other aspect of the room acoustics, whether it's side walls, back walls, and ceiling, it seems like a lot of people have strong belief at what their saying, yet when I read it all, it just sounds like one big contradiction, and I'm left with the feeling that maybe it will be just better to not even try to start with it.

Don't get me wrong, I know no one can tell me for sure what is going to to work perfectly in my room, but I'm not looking for this, all I'm looking is for some ideas about areas that there are high chances that needs absorption apart from first reflection points.

I read one guide (a pretty comprehensive "101" guide from one of the websites, I think it was "Auralex") which said that for HT usage, you should try to deaden all the room to the same extent, creating an even "ring" of tiles all around you, and even if it is going to ruin the stereo experience, it should be, hands down, an ideal situation for HT.

One of the things which crossed my mind, is to refuse to play this "guessing" game, and just create this 3-4 feet high "ring" across my small room, centered at ear level, so all frequencies from ~80hz to 20khz will get minimal reverberation time. Below 80hz, the digital room correcion will balance things very efficiently, so at the end, the entire frequency spectrum will have a pretty similar reverberation time, which will be pretty short. This should be optimal for HT movies and Dts/DD/DVD-A/SACD 5.1 music CDs, where wall reverberation (except ceiling, since there is no ceiling speaker) has little meaning (since they are reprodiced by the surround sound channels), and as for stereo, well, I'll just give up the idea of an acurate stereo presentation, and switch DPL-II, LOGIC-7 or Circle Surround II, to try and mimic the decay of sounds with my surround speakers. At least in this way I'll be in control of my room, instead of the opposite...

 

[Jones_Rush]

edit.

 

[Pat]

Jones, glad my step by step process is helpful

 

[Jones_Rush]

Yes Pat, very! .
Btw, is there anything about the construction you would have done differently, if you were to build it again today ?. I saw you didn't use a layer of poly batting behind the fiber glass (only in front), is there a reason why not ?.

 

[Jones_Rush]

Here are some waterfall images of the left speaker, at the listening spot. This is with digital room correction (which can't do a good job for the higher frequencies reverberation anyway) and with acoustic treatments (heavy blankets and pillows at first/second reflection points, and at the closet behind my head).

1ms:
http://www.avsforum.com/avs-vb/attac...chmentid=12505

3ms:
http://www.avsforum.com/avs-vb/attac...chmentid=12506

5ms
http://www.avsforum.com/avs-vb/attac...chmentid=12510

10ms
http://www.avsforum.com/avs-vb/attac...chmentid=12509

The situation is very similar with the other speaker too.

As you can see, up to about 2ms, the work of the absorbers can be seen (the speakers are 1 feet away from the side wall next to them, no option to change that). After that, things become really bad, really fast. Some say that in order to get good sound you need at least 5-10 ms of no reverb after the main spike. There is no possible way to achieve this, as you can see, without going with a much more comprehensive absorbing scheme, which might need to include an entire absorbing "ring" across the room. Treating just points simply won't work.

 

[ThomasW]

About the "clapping" test, right after the clapping, I hear a faint but noticeable metalic, high pitched echo, on both ears. I have no clue how it in practice affects the sound of the music, but anyay, it's definitely there with the clapping.

That's slap echo and it will have a negative impact on all the sound.

If your listening position is against the back wall/closet you'd definitely benefit with some thick panels there. 10" is a good depth there if possible.

The highest octaves of the Behringer can't be trusted. we compared one to a B&K 4133. The match was actually pretty good up to ~10-12kHz where the 8000 started rolling off.

Even though I'm a 'geezer' I can't tolerate any form of digital EQ. All but the $$$$$$ (think Tact) models sound like finger-nails on a black board.

 

[Jones_Rush]

Even though I'm a 'geezer' I can't tolerate any form of digital EQ. All but the $$$$$$ (think Tact) models sound like finger-nails on a black board.

I don't know if my room correction system if better or even as good as the Tact, but it definitely doesn't sound like "finger nails on a black board". According to someone who presumably compared both the room correction implementation I use, and the TacT RCS 2.2X (he owns the Tact), in his own system, he said that he is not sure if he can detect a difference. This was 10 months ago, since then the room correction software I use got much improved. (This guy's name is Patrick, and he even dedicated a website to the room correction software I use, he also say there that he is going to compare the TacT 2.2 to DRC, but he never updated the guide after he did the test. I know what he thought from a different, but trustworthy source).

Anyway, you can try it for yourself, for free, given you have some basic measuring parts, and a decent sound card.

DRC Guide

I wish more people with the TacT RCS will try this room correction option (with their own mic), and give their opinion.

PS.
What you say about the ECM8000 is interesting. After I'll finish with the room treatments, and be sure my sweet spot is arranged exactly the way I want it, I might rent a higher quality mic for a day, in order to create filters which will be accurate also above 12khz.

 

[RichardHOS]

Jones,

There is no possible way to achieve this, as you can see, without going with a much more comprehensive absorbing scheme, which might need to include an entire absorbing "ring" across the room. Treating just points simply won't work.

This is actually the room treatment approach favored for a multichannel setup. A ring of absorption around the room at ear level, and more absorption where needed to achieve the desired RT60 times, starting with first reflection points (by adding material above/below the ring at those points) and the front wall (behind speakers), and then continuins with other reflection points if needed.

For two channel, well, I think it's nearly impossible to have the appropriate RT60 and a well behaved impulse out to ~10ms unless the room is really large. Extensive diffusion is about the only chance there. In general, setting up the room for proper multichannel is a much more straightforward task than setting it up for proper two channel, which is why I'm among those that advocate the use of digital processing of two channel source material to create the proper reverberant field rather than the use of the room as a processor to achieve the same, but perhaps not as successful, end result.

 

[Jones_Rush]

RichardHOS,

Your post was all that needed in order to push me to the dark side, completely . Goodbye 2ch stereo, Hello multichannel! (and huge, HUGE amounts of Miraflex. )

 

[Jones_Rush]

Richard,
I'm wondering, do you think that DPL-II or Circle Surround II in multichannel with a deaden room, will be at least as good as a typical LEDE room with plain 2ch stereo ?. I have access to both these algorithms through the PC (I'm running my sound system through the PC).

 

[RichardHOS]

I don't have any direct experience with CS or CSII. I have heard a great L7/dead room implementation, and no stereo setup I've ever heard could compare. I'm not sure how CSII and L7 stack up, other than the anecdotal knowledge that L7 is highly respected.

I will say that a good LEDE implementation can sound fantastic, that much is sure. A good multichannel implementation can also sound fantastic... of that I am also sure. The important piece of the puzzle is that LEDE isn't so great for multichannel playback, which makes it incompatible with movie and multichannel audio uses to a greater or lesser degree (depending, of course, on how live the LE is and how dead the DE is).

So... with plain 2ch stereo I think it would likely be a tossup with between a good LEDE room and a dead/multichannel setup using CSII, DPLIIx, etc. L7 and Trifield would likely be better, all other things being equal (source, speakers, etc.).

 

[Pat]

Jones,
The fiberglass batts have a paper backing on them, so putting any dacron behind wouldn't do much good.

I basically followed Thomas' directions except that I used burlap instead of muslin.

 

[Jones_Rush]

Richard,
How important is the center channel when DSP'ing a stereo signal, to a multichannel one ?. Is it good practice to go with only two speakers at the front, and four at the back ?.

For music alone, I never find the center channel too useful. For movies though, it's an entirly different story.

 

[RichardHOS]

That depends on the quality of the DSP, and the quality of your center channel. Typical centers with limited extension and poor horizontal dispersion aren't exactly to best choice for music.

However, with a good center (i.e., same as L/R ideally) and good DSP (trifield, L7), there are some very good reasons for not removing it from the system. Reduced comb filtering will result in an increased sweet spot size and image stability. Just remember... when you hear a trumpet solo on stage, it comes from a point source. We (audiophiles) have for years had this concept that stereo (two-channel) is the "proper" way to recreate the front soundstage, and that a center channel messed that up. Well, that's just a result of having grown up (most of us) with two speakers. Go listen to a real symphony and you'll immediately see where a center channel comes into play.

Personally, I hope we eventually wind up with something like 5 horizontal discrete channels up front (and maybe a couple more for height, plus whatever rear concoction they come up with).

That being said, not all centers are created equally, and (from my admittedly limited time with L7/trifield) not all DSP's are created equal either.

 

[Jones_Rush]

In my case, the major problem is the center location. I simply can't put it at the same height level of the mains, or bring the mains lower.

Somehow, and I really don't know how to explain this, for movies the imaging really lacks without the center speaker, everything sound MUCH better with the center (the room is practically energized with the center, and completely drown out of life, without it). But, with music, the inaccuracy of the center's height, really comes into play. I get much better imaging without it (even with pre recorded 5.1 music).

Anyway, according to what you say, it doesn't seem like I'm going to lose too much by going with a phantom setup, if I'm satisfied with the image stability and wideness of my current stereo configuration.

 

[Jones_Rush]

What options do I have, if I don't want to go with fiberglass at all (not even Miraflex) ?. Isn't there any safe (health wise) alternative which absorbs good enough, even for more money ?.