What's new

Odd room response measurements (Kit281) (1 Viewer)

Michael R Price

Screenwriter
Joined
Jul 22, 2001
Messages
1,591
I borrowed a SPL meter from a friend and measured the bass response from 25-200 Hz 1/3-octave, at my seating position in 3 configurations: 1) with only the left speaker, in its normal position; 2) with only the left speaker, sitting in and pointing towards my front left corner; and 3) with both speakers (mono) in their normal positions. Oddly enough, the response declined quickly below 200 Hz (20 dB or more by 25 Hz) whereas LspCAD predicted +/- 4 dB from 20 Hz up.
Here's the Excel graph:
kit281bass.gif

Weird, eh?
I had been noticing a few things before the measurement:
- The bass sounded very natural and deep, but there was less of it than I was used to. I thought this was because the 281s deliver accurate and non-boomy bass. But it honestly didn't seem like 10-15 dB less than normal.
- Walking around the room created much more bass output but it sounded 'uneven' and boomy, like certain frequencies got exaggerated a lot more than others.
- Bass that was hair-raising and 'chairshaking' didn't seem as loud as the physical movement would indicate.
So I'm really not sure what is going on here, since even corner loaded there was a big decrease in output. I'm very hesitant to doubt the meter (RS analog, C weighting slow response, rolloff compensated).
The question is:
Are the speakers a problem? I have tried to seal off any air leaks (my construction on the 281s was not very good) but there may be a few minor ones still. (Would that cause such a dramatic loss of output?)
Is my seating position, and not speaker position, the problem? I think so, because the results from corner loading were similar to the others. But I'm not sure how one particular spot in the room can just cancel out everything like that.
What else? I felt like I was getting really great bass out of these things but now it seems as if I'm *seriously* missing something.
The speakers face down the long way of my 15x11x8 room, 4 feet from the back wall and 2 from the sides. I sit between the speakers 7 feet from them and 4 feet from the back wall...
Is there any way to calculate the room response based on a position in a room? I've heard about dividing room dimensions into 565 to get node frequencies... but this seems like a bigger problem. I'm quite confused. :frowning:
Thanks for the help, guys. I greatly appreciate it.
 

Dan Wesnor

Second Unit
Joined
Apr 28, 1999
Messages
389
The 15' dimension of your room is creating a null at about 37Hz and its harmonics, which is clearly visible in the plot. The room mode at 11' (52Hz & harmonics) appears to be excited, which creates a lumpy graph. This is normal for a small room, not much you can do about it. Your best bet is to offload the bass to a subwoofer, which you can place so as to minimize the effect. Or get a really, really big room. :)
Leaks can affect the tuning of the box, which can affect the level and quality of the bass.
 

Fred Seger

Stunt Coordinator
Joined
Aug 13, 2000
Messages
136
could it be your amp??? some amps don't do as good as others with bass. Tone controls on the receiver set balanced??

Let us know what you find.
 

Michael R Price

Screenwriter
Joined
Jul 22, 2001
Messages
1,591
Thank you for the replies.

Dan, I understand the standing wave situation but what would cause the overall downward slope of bass response? Down 20 dB where it would theoretically be only 2-4? Too bad I can't afford a subwoofer.

Fred, I think the amp is fine. I ran the tests at fairly low power levels (75 db) and I doubt it would cause such large variations.

What I'm worried about is needing to change my seating position. Would moving the speakers affect response at a certain spot, or does each position in the room have its own response regardless of speaker placement?
 

Greg Monfort

Supporting Actor
Joined
May 30, 2000
Messages
884
Assuming you have no EQ, teeny gauge speaker wires, or amp problem, the only things that I know of that causes this much roll-off in a corner is a leaky cab or driver/speaker wired out-of-phase. Since these have dual woofers, did you purchance accidently internally wire one up wrong, and for sure verify that all cab joints/driver mounting are sealed.

Rooms have basic modes and how much they're excited/modified is dependent on speaker positioning, and of course where you sit is some different set of modes so getting a good response at the listening position is a balancing act.

GM
 

Dan Wesnor

Second Unit
Joined
Apr 28, 1999
Messages
389
First of all, an AV-8 won't do 20Hz, so I'm thinking LspCAD was doing an in-room simulation, which is a swag at best. An AV-8 is only good down to about 40Hz, unless you have a bad alignment in a huge (200L+) box.

When I as in a smaller room, I had 15dB swings from nominal in my sub's response. I think you may be seeing a 10dB dip right at the corner frequency.

Whenever you do bass measurements, you should take one measurement well out of the range of room modes (300Hz for a sub, 1kHz for a full range), to determine what the nominal sound level is. From your plot alone, I can't tell whether that's a 20dB dip or a 10dB dip next to a 10dB peak.
 

Michael R Price

Screenwriter
Joined
Jul 22, 2001
Messages
1,591
Dan, good point about getting a 'reference' SPL reading. I've got to borrow the sound meter again (don't have one myself but a friend does and is willing to lend it to me). I feel bad for posting all this stuff when I left out such an important detail. But this is still an interesting discussion, to me anyway. I'm learning a lot as I've started to mess around with my setup more recently.
Here's the LspCAD sim.
kit281sim.gif

Mine are the standard 61 liter vented box tuned to a lower than usual 24-25 Hz. (I did this based on LspCAD's room response predictions which may very well be incorrect. Oh well.) So they are down about 10 dB at 25 Hz.
Is it possible for air to leak out through the driver? A couple spots on my woofer surrounds were loose and caused a buzzing/flapping noise and also an air leak. I superglued them to the AV8 frame and that seemed to stop the problem... also, I did not use the piece that came with the driver (the mounting gasket). At the time I thought it was just a piece of cardboard used to protect the driver in shipping. (Oops again.)
Is there any software that can show you the bass response at any point in a room (accurately)?
If you divide 565/d where d is a room dimension, are the odd-order (1,3,5,7x) frequencies nulls or peaks (I asssume the even order frequencies are the opposite)?
Greg, they're not out of phase. All 4 woofers total both move in sync when I play a 5 Hz tone. Thanks for reminding me though.
Maybe the problem is not as bad as I thought. One side of me says, music sounds great so what's the big deal with your measurements? The other side says, what the heck is going on here?
 

Jacques C

Stunt Coordinator
Joined
Nov 28, 2001
Messages
84
Michael,

I have 281s, and also happen to have a simulation setup using CARA for 281 in-room response. I ran your room and seating position through it and I find:

A dip at ~50Hz. yep - you measured it.

A dip at ~75 Hz. Yep again.

Not sure about your dip at ~200 Hz.

I can see only two causes to your problem. First - out of phase drivers. I think you crossed that one off. Second - what are your current bass management settings? Do you by chance have your speakers set to small and crossing over to a non-existant sub at 80 or 100 Hz? That would certainly do what you are experiencing.

BTW, your "SPL room and cabinet" is very close to what I calculate. You should not see a huge rolloff starting at 100Hz by any stretch - and I can tell you that my 281s run "large" easily hit 40Hz flat in-room. Those AV8s most definitely do not roll off at 100 Hz. ;-)

Not sure if it helps, but hopefully it will a bit.

Jacques
 

Dan Wesnor

Second Unit
Joined
Apr 28, 1999
Messages
389
First of all, I'd raise the tuning to 37Hz or so. At 24Hz, the port isn't really doing much.

Whether you get a null or a peak depends on the listener location. How much of a null or a peak you get depends on speaker placement and listener position. If the speaker sits in a peak, it excites the room mode. If it sits in a null, it gets canceled by the room mode. By the way, this happens in 3D, so vertical placement of the meter can also affect the results.

In my experience, attempting to solve these problems in a small or complicated room can be frustrating if not futile. In my old house, moving six inches in one direction greatly affected bass. It was a 12x20 room that opened up into 2 other rooms which acted as perfect resonance chambers.

And the LspCAD plot shows the null near 200Hz that shows up in the measurement.

I'd redo the vents, not measure again, and live in blissful ignorance. At most I'd EQ out the hump, but that's about it.
 

Michael R Price

Screenwriter
Joined
Jul 22, 2001
Messages
1,591
And how exactly would I modify the vent? The cabinets are glued shut...

Sealing up the vent wouldn't work because the cabinet is too big (it would be a super low Q sealed box and have even less output where it's lacking).

Guess I'll have to live with this in non-blessful ignorance for now. Again thanks for the help.
 

Greg Monfort

Supporting Actor
Joined
May 30, 2000
Messages
884
>Is it possible for air to leak out through the driver?

====

Yes, but not in a design like the AV8 if properly glued together.

====

> A couple spots on my woofer surrounds were loose and caused a buzzing/flapping noise and also an air leak. I superglued them to the AV8 frame and that seemed to stop the problem... also, I did not use the piece that came with the driver (the mounting gasket). At the time I thought it was just a piece of cardboard used to protect the driver in shipping. (Oops again.)

====

But you did seal the drivers to the baffle, right?

====

>Is there any software that can show you the bass response at any point in a room (accurately)?

====

Don't know of any.

====

>If you divide 565/d where d is a room dimension, are the odd-order (1,3,5,7x) frequencies nulls or peaks (I asssume the even order frequencies are the opposite)?

====

All modes are nulls since the baseline is flat.

====

>Greg, they're not out of phase. All 4 woofers total both move in sync when I play a 5 Hz tone. Thanks for reminding me though.

====

Ok.

====

>One side of me says, music sounds great so what's the big deal with your measurements? The other side says, what the heck is going on here?

====

I'm with you. Something's not jiving. Even after your remark about lower tuning (not much lower than the kit recommendation if you figured it right) and the major difference I found between the spliced together response on Adire's site and what I simmed using the info from the kit plans, the corner loaded FR still doesn't make any sense to me unless there's a problem or the corner has a doorway or other cutout near it.

GM
 

Users who are viewing this thread

Forum statistics

Threads
356,818
Messages
5,123,879
Members
144,184
Latest member
H-508
Recent bookmarks
0
Top