Room Analysis Results

[redrue]

I've built the bass traps and installed them in my
mix room. I used Auralex 4" mineral fiber in 1X4
pine frames covered with muslin. They're 2X4'
installed 4 in ceiling/wall corners and 4 in corners...
I also have Auralex Max Walls up. There are 4 panels
of 703 on the ceiling.

I borrowed an Audio Analyzer (Sencore SP295) and
I've graphed the results of the Real-Time Analysis.
Ran test @ 95db SPL with Pink Noise 1/12 octave...
Room size is 13 X 11 X 8h

I've posted the graphs in three separate files...(only
way I could)

It seems to me that my 400-6300 is relatively flat... the
low frequencies have a couple of dips between 50-72hz and
89-125hz. The high end keeps a pretty steady increase trending
pretty much in line with the slope in the 'mid' graph.

What I'm wondering here is how good is this compared to
what you're familiar with? Can these dips in the low freq.
be flattened out with more traps... do they need to be flattened
out? What about the upward trend in the higher frequencies? Is
that 'normal' or do I need a little more HF treatment?


Appreciate any advice,

thanks

Rue

p.s. The room sounds great to me - a huge improvement. I
just want to make sure after I've come this far that I get
the job done as well as possible.

[Ethan Winer]

Rue,

> Ran test @ 95db SPL with Pink Noise 1/12 octave...
> Room size is 13 X 11 X 8h

1/12 octave is reasonably accurate, and certainly much better than 1/3 octave, but you need an even finer resolution to see the true extent of the peaks and nulls at low frequencies. Also not shown in frequency response graphs like these is low frequency ringing which is at least as important as the raw response.

> Can these dips in the low freq. be flattened out with more traps. <

Yes, that's exactly the right way to do it.

> The room sounds great to me - a huge improvement. <

Yep, and this proves that you're headed in the right direction.

--Ethan

[Glenn Kuras]

From looking at your graphs things are looking pretty good, but I see no reason why it could not be better.. You may want to build maybe 4 more traps and run the test again.
Also there is a program called ETF you may want to look into... It will give you a much better resolution and also so you what the RT60 is.. Very important as Ethan pointed out also.. I believe the website is www.eftacoustics.com.

Glenn

[Cojo]

Originally Posted by myfipie

From looking at your graphs things are looking pretty good, but I see no reason why it could not be better.. You may want to build maybe 4 more traps and run the test again.
Also there is a program called ETF you may want to look into... It will give you a much better resolution and also so you what the RT60 is.. Very important as Ethan pointed out also.. I believe the website is www.eftacoustics.com.

Glenn

You got the link wrong there...

Try this instead: http://www.etfacoustic.com/

/Cojo

[macr0w]

That eft site looks interesting. I would love to shoot my room and see how bad it is. I used a hardware device that I borrowed from a friend a while back(can't remember the name). It used a 1/3 octave led display, not very accurate. Do you guys think it's worth it to get their measurement mic for $325.00 with this software? Or could another mic be used?

[mltamisin]

River Audio offers a mic/pre combo for $250.00. I don't know the difference though.

[Ethan Winer]

Mac,

> Do you guys think it's worth it to get their measurement mic for $325.00 with this software? <

The biggest problem in all rooms is the low frequency range, and even a $50 Radio Shack SPL meter is fine for that. But for higher frequencies a better mike is a good idea.

Also, a huge feature of ETF is it shows modal ringing over time, which is at least as important as the raw response if not even more so.

--Ethan

[ianfreefall]

Hi Ethan

I've got quite a bit of 64kgm3 fibreglass damping in my room already - 2'x8' (6") over all the wall/wall corners, some 4" behind the monitors and on the back wall. (Also some 2" on the side walls and ceiling, but that's to tame the high end first reflection positions). It's all spaced carefully, and the frames are open at the sides etc to get best performance.

The room is a concrete box/hardwood floors: 16.5' x 12.2' x 9' - not great measurements, but better than some. I've got the mix position right - 37% from the front walls, slightly off centre, monitors firing down the long dimension. Good quality monitoring (Genelec 8030, with a sub coming, hence the steep dropoff below 60hz).

I've now tested the room with your test tones - 1hz spacing - but I'm on a Mac and don't have virtual PC so I haven't run the ETF software yet - so no modal ringing tests...

But there is still a fairly large boom around 66hz - repeating at 125hz, and then less so at 180hz. I haven't bothered to calculate, but I guess this is the standing wave dimensions of my room.

I've picked up more fibreglass and drilled in the hooks already to treat to the wall/ceiling corners... is this heading in the right direction? How does this graph look to you - a 26db range seems like a hell of a lot to me, and I expected better...

http://www.ianhenderson.co.za/TestRoom.pdf

[Glenn Kuras]

>I've picked up more fibreglass and drilled in the hooks already to treat to the wall/ceiling corners... is this heading in the right direction?<

Without seeing what you had before it is hard to know how much better it is, but it sounds like to me you are heading in the right direction.. With a room like yours you are going to need A LOT of trapping do to the concrete walls... We did a room that had concrete on the floor, walls and ceiling and I think we ended up putting 14 traps in the room!

Glenn

[ianfreefall]

That's what I thought - I've already got 13 x 4" and 6 x 2" traps in place, and I'm planning another 7 across the ceiling/wall corners!

[Ethan Winer]

Ian,

> a 26db range seems like a hell of a lot to me, and I expected better... <

I'd have expected better too, but there are a lot of factors like how rigid and dense the walls and ceiling are. Without seeing what you had before, or how the modal ringing was and now is, it's difficult for me to comment. I have to assume that with all four corners treated at least it sounds better than without the traps, yes? Regardless, treating as many wall-ceiling corners as possible can only help further.

--Ethan

[ianfreefall]

Thanks. Well, I'm going to try and get some modal ringing information. It's a damn rigid, old block of flats - serious concrete everywhere. I didn't do any stats before I started because I knew they'd be horrible. But I'll get the extra traps up and then retest and see where we are.

When I'm fully treated, and still having big peaks at 60/120hz, would it be worth patching eq over the mix bus?

thanks
Ian

[Glenn Kuras]

Originally Posted by ianfreefall

Thanks. Well, I'm going to try and get some modal ringing information. It's a damn rigid, old block of flats - serious concrete everywhere. I didn't do any stats before I started because I knew they'd be horrible. But I'll get the extra traps up and then retest and see where we are.

When I'm fully treated, and still having big peaks at 60/120hz, would it be worth patching eq over the mix bus?

thanks
Ian

I would not eq over the mix bus!!!!!!!!!!!!!!!! You may think you are solving one problem but you could be f'ing up something else... I really think if you take the time to build more traps and get them in the corners you are going to solve the problem... You may want to think about making a few of them 6" thick and start with those in the front corners of the room...

Glenn

[ianfreefall]

Well, actually, the traps over the four wall/wall corners are 2' x 8', and are 6" thick. So I'm going ahead with the wall/ceiling corner traps and we'll see after that.

With the corner traps is it particularly important to get them flush with the walls, to "seal" in the corners? I've got a concrete beam running down the top of the one side wall that protrudes 6", and so there's quite an air gap around the trap edges - it's across the corner, but not quite in the corner, if you know what I mean.

I must say, on the whole the room does sound a lot better, even if it still booms like mad when that 90db test tone at 120hz is vibrating around the room.

[Glenn Kuras]

Originally Posted by ianfreefall

Well, actually, the traps over the four wall/wall corners are 2' x 8', and are 6" thick. So I'm going ahead with the wall/ceiling corner traps and we'll see after that.

With the corner traps is it particularly important to get them flush with the walls, to "seal" in the corners? I've got a concrete beam running down the top of the one side wall that protrudes 6", and so there's quite an air gap around the trap edges - it's across the corner, but not quite in the corner, if you know what I mean.

I must say, on the whole the room does sound a lot better, even if it still booms like mad when that 90db test tone at 120hz is vibrating around the room.

I think I understand what you are saying and no not having it sead is fine.. The more gap you can have the better off you are.. There is a point of to much, but I think you are fine..

Glenn

[Cojo]

Originally Posted by ianfreefall

Thanks. Well, I'm going to try and get some modal ringing information. It's a damn rigid, old block of flats - serious concrete everywhere. I didn't do any stats before I started because I knew they'd be horrible. But I'll get the extra traps up and then retest and see where we are.

When I'm fully treated, and still having big peaks at 60/120hz, would it be worth patching eq over the mix bus?

thanks
Ian

In that case I would rather patch the eq on the monitor path instead of the mix bus.

/Cojo

[tINY]

The 66/132 Hz ringing is from the walls that are about 17 feet apart. There is probably a 33 Hz ringing too, but a lot of speakers wont really excite this.

I see the littlw peak at about 99 Hz too.

Concrete is a very ridgid wall. Building another wall of gypsum board in front of it and filling the cavity with insulation might help. Of course, this is just a panel resonator trap (like Ethan describes on his site).

While you are at it, angling the walls or panel traps at more than 10 degrees might help too. Make sure that you make it thicker in the middle and less on the sides (convex).



-tINY

[Ethan Winer]

Ian,

> would it be worth patching eq over the mix bus? <

No. There are a lot of problems with EQ. For starters, the peaks and nulls vary all over the place in the room, even across a span of just a few inches. So as soon as you add EQ to make the response better in one place, it will likely be worse somewhere else. I don't mean that you can fix the response at the mix position but the couch in the back might be worse. I mean that even leaning forward a few inches will make things worse. Don't do it. Treat the room as best you can and accept the results.

--Ethan

[Glenn Kuras]

Originally Posted by Ethan Winer

Ian,

> would it be worth patching eq over the mix bus? <

No. There are a lot of problems with EQ. For starters, the peaks and nulls vary all over the place in the room, even across a span of just a few inches. So as soon as you add EQ to make the response better in one place, it will likely be worse somewhere else. I don't mean that you can fix the response at the mix position but the couch in the back might be worse. I mean that even leaning forward a few inches will make things worse. Don't do it. Treat the room as best you can and accept the results.

--Ethan

Also you can not eq a null...

[anahata]

Originally Posted by myfipie

Also you can not eq a null...

&lt;giggle&gt;
Having seen the nulls you can find with Ethan's 1Hz-spaced test tones, I sometimes wonder if the "fix it with EQ" enthusiasts have any idea what they're up against. Also I found (as expected) *totally* different response patterns in two measurements, one with the mic in the middle of the room and the other right close to the back wall. Great big nulls both times, but at different frequencies.

In order to find out the best place for absorption (ends/sides/corners etc) it might be worth setting up a tone at a frequency where you already know there's a peak or a null and wandering around the room with a mic watching the meter to see how the pattern varies. You should soon get a feel for whether it's an end-to-end, left-to right, top-to-bottom or corner-to-corner thing.

[Ethan Winer]

anahata,

> I sometimes wonder if the "fix it with EQ" enthusiasts have any idea what they're up against. <

No, but they sure think they do!

> one with the mic in the middle of the room and the other right close to the back wall. Great big nulls both times, but at different frequencies. <

I've measured big changes even a few inches away.

--Ethan