Diffusion v. Absorption? Room size?

[allredp]

Newb here...

Among general acoustical issues for my room, I need to understand what the difference is between diffusion and absorption! I've read in several different places that room size dictates which (or both?) you want to use.

My room is 27' W x 15' D x 92" H--I have a 6' angle wall in the back L corner with a gas fireplace/mantle in it. There is also a 4' x 2' closet that outcrops in the back R. I've dropped three pics in so you can see what I'm describing.

My initial tests of the room showed a -9db down around 16kHz, as well as -9db down between 260-300 Hz. This was using a 31 band spectrum analyzer with 3 db increment LEDs. I have a friend coming Tuesday with a laptop to get more precise FQ measurements.

I'm quoting from Bryan's helpful thread on general definitions for acoustic treatments at the beginning of this forum:

"Generally, you're best off if the distance from speaker face to front wall, driver centers to side wall, and driver center to floor are 3 different dimensions in order to not reinforce any specific set of harmonics by having all the boundaries generate the same SBIR effect.

My mains are 28" off the front wall and 100" off the R and 10' off the L side walls. The drivers are anywhere from 38" to 16" (tweeters to woofers).

So, with a room this size and with my mains and SVS PB 12 Plus/2 placed where they are (85" apart, toed-in slightly, and 100" from primary listening position) do I need diffusion or absorption?

Thanks for any help!
Phil

[bpape]

Absorbtion.

The diffusion would be useful if the room was oriented to have the speakers firing down the long dimension. Diffusion would be on the back wall (behind your head). In this orientation, the best you could hope for in the way of diffusion would be for the side wall reflection points. To deal with broadband diffusion, they'd be probably 7-8" deep.

Being that far down at 16kHz is not room related - guaranteed. It's a function of the mic or something else. As for the bottom end, I'll wait and see what the fine measurements show - please try for no more than 1-2Hz coarseness in the plotting.

Almost certainly, broadband bass absorbtion and some additional soft decay time control/reflection control is the solution for this space. The one thing you do have to conern yourself about is the depth of your modes. The long room dimension will put some of the axial modes down in the 20's which are very very hard to deal with unless you have room for something a couple of feet thick. Most people prefer to leave that range alone anyway for a little extra 'oomph'. We'll likely concentrate from the first multiple of that mode and up.

Bryan

[allredp]

Excellent, Bryan!

I've downloaded some test tones at 1Hz intervals (up to 300Hz) and will plot them a.s.a.p.

I just need the mic (RS SPL meter) at listening position and the tones playing through a CD player, right?

Is there a slick way to convert my single point hand-drawn graph to an electronic plot?

Thanks so much for the clarification and help!

Phil

[bpape]

If it was me, I'd take a sideways click here and find the Room EQ Wizard area. The tool is free, you can use the RatShack meter as the mic, you can use a plug-in to compensate for the meter/mic inaccuracies, etc. It'll run an entire sweep and keep things in very tight resolution. It will do all this in about 10 seconds so you can easily play and run multiples.

The other advantage is that it will give you not only frequency response, but also decay time/waterfalls, and impulse response.

Bryan

[allredp]

OK, Bryan. Good idea, though I haven't the foggiest about the REQW! I'll be studying about how to do it tonight...

One question: you mentioned mic calibration issues (which I have heard about in other areas). The fix for that you said was a "plug-in" that would compensate for the meter/mic inaccuracies.

Please help me understand what this "plug-in" is and where I find it!

Is is part of REQW? or something/somewhere else?

I really, really appreciate this help!!!

[bpape]

If you look at the 'how to' thread for REQ, I think there is a mention of a calibration file that can be selected from within REQ that will compensate for the mic inaccuracies. It's not perfect as every mic is different but it will get you pretty close.

Bryan

[altenburgpj]

I would consider absorption on the back wall that extends from the seating to the ceiling. Other than that, your distance to the sidewalls is probably far enough that unless there is either a low frequency mode that is problematic or an excessively long reverberation time, treating them would be more to your personal taste than a requirement.

Send some graphs when you have them.

[allredp]

Excellent ideas--WAF will dictate what I get away with in this room!

I had my tech friend over last night with his test gear (a laptop with stereo sound card, a test mic or some kind with a phantom power pre-amp, and a full-range pink noise cd played through my CD player/AVR in stereo mode = mains + sub). The RTA program he was using was called SoundLive I believe--the program had the look of the other brands of RTA I've seen posted here. He grabbed the paused graphs with the screen copy function and will be sending me the files later today so I can post them.
However, when we isolated the lower FQ (16Hz--300Hz), the 1/24 only registered some select bands. In fact, everything below around 100Hz, if I remember right, was really far apart. I could only see one bar (or line point) for 63, then 31.5. Strangely it was acting like a 31 band analyzer below 100 Hz or so! Very strange I'm afraid, and probably rendering our work last night unhelpful.

I'll most likely have to learn REQW for myself, eh?!

At any rate, I'll post those graphs (screen shots) later today to see what you think.
Thanks for all,
Phil