You are here

Log in or register to post comments
Buddha
Buddha's picture
Offline
Last seen: 2 years 8 months ago
Joined: Sep 8 2005 - 10:24am
Measurment question...

Aloha,

You know how it is, February 14th, Cupid's Day, a time when spring is in the air and a man's thoughts take flights of fancy as he ponders...hi fi.

So, anyway, I gots a question:

I see the speaker measurements JA does for reviews and it makes me wonder if we could actually end up plotting/inventing a room's frequency response, as well.

Let's say we take all his open air measurements and then compare each speaker's curve to his in-room measurements, and then subtract the differences.

Over a period of time and speakers, as you refine this difference curve, could you say that you've come up with a sort of real world frequency response correction for his room?

Then, would that be the same curve as he would get by using a digital room correction program?

Which leads me to my next question...

Wouldn't all this make it possible to measure the impact of different room tunes, etc, and make for a neat survey of the effects of room mods?

I can't recall ever seeing, say, what a good set of bass traps do to a given room. I'd be interested. Has anyone else seen anything like that? I wonder what my couch may be doing, too.

I'm not trying to reduce everything to measurements, just trying to think of what other cool things we could see to chat about with these machines.

Kal Rubinson
Kal Rubinson's picture
Offline
Last seen: 18 hours 2 min ago
Joined: Sep 1 2005 - 9:34am
Re: Measurment question...

Good points. The application of bass traps is very room-specific and position-specific but such measurements would distinguish between those that actually have an effect and those that do not. For a case study, you might look at:
http://www.sbrjournal.net/currentissue/articles/acoustics/Acoustics.htm

Kal

Buddha
Buddha's picture
Offline
Last seen: 2 years 8 months ago
Joined: Sep 8 2005 - 10:24am
Re: Measurment question...

That case study was fantastic!

That one will be read over and over. Very info-dense!

Thanks.

Jim Tavegia
Jim Tavegia's picture
Offline
Last seen: Never ago
Joined: Sep 1 2005 - 4:27pm
Re: Measurment question...

One of the things I have done in the past over the years is use a measurement mic that is considered pretty flat and do what JA did when he was trying to measure the actual freq response with the large Bozak's. 225 lbs is alot to drag around.

I put the measurement mic at the listener position and run white noise through the system and record it into my laptop and view the results in my recording software. It may not be perfect, but I feel like it is telling me something about how the speakers are reacting to the room and what the result is. I doubt the LF info is terribly accurate, but I can usually use the "Phile" freq sweeps from the test CDs and get a sense of any sudden peaks or valleys from 500 hz and down.

Editor
Editor's picture
Offline
Last seen: 3 years 7 months ago
Joined: Sep 1 2005 - 8:56am
Re: Measurment question...


Quote:
Let's say we take all [JA's] open air measurements and then compare each speaker's curve to his in-room measurements, and then subtract the differences.

Over a period of time and speakers, as you refine this difference curve, could you say that you've come up with a sort of real world frequency response correction for his room?

Something like this has been on my to-do list for a long time. I believe that you would be correct, that the overall difference curve would be the inverse of the optimal DSP correction curve. However, one confounding factor is that my room responses are limited to 1/3-octave resolution.

John Atkinson
Editor, Stereophile

Kal Rubinson
Kal Rubinson's picture
Offline
Last seen: 18 hours 2 min ago
Joined: Sep 1 2005 - 9:34am
Re: Measurment question...


Quote:
Something like this has been on my to-do list for a long time. I believe that you would be correct, that the overall difference curve would be the inverse of the optimal DSP correction curve. However, one confounding factor is that my room responses are limited to 1/3-octave resolution.

John Atkinson
Editor, Stereophile

I've been thinking about this (and fooling around with it) for a little while. I doubt if one can define that room response independant of speaker position and listener (microphone) position. If you controlled these tightly, you might be able to get a curve that would permit you to get some meaningful in-room measurements as long as you don't change any furniture or move around any accumulated materials.

Another issue might be the existence of very high q nulls or peaks which would swallow up whatever the speaker is doing at that point.

I am happier with out-door measurements but, then again, I am not the one doing all the heavy lifting.

Kal

Editor
Editor's picture
Offline
Last seen: 3 years 7 months ago
Joined: Sep 1 2005 - 8:56am
Re: Measurment question...


Quote:
I've been thinking about this (and fooling around with it) for a little while. I doubt if one can define that room response independant of speaker position and listener (microphone) position. If you controlled these tightly, you might be able to get a curve that would permit you to get some meaningful in-room measurements as long as you don't change any furniture or move around any accumulated materials.

Right, I didn't mean to imply otherwise. When I perform my published in-room curves the 10 mike positions are controlled with an error of

kfalls
kfalls's picture
Offline
Last seen: Never ago
Joined: May 30 2006 - 8:53am
Re: Measurment question...

Don't forget, depending on the meter you're using there may be a correction curve for meter itself. I believe the Radio Shack analogue meter is 4db down at 30Hz. I believe the bass has the largest deviation from flat response. Search the net for the correction curve.

Editor
Editor's picture
Offline
Last seen: 3 years 7 months ago
Joined: Sep 1 2005 - 8:56am
Re: Measurment question...


Quote:
Don't forget, depending on the meter you're using there may be a correction curve for meter itself. I believe the Radio Shack analogue meter is 4db down at 30Hz. I believe the bass has the largest deviation from flat response. Search the net for the correction curve.

I don't use the RadioShack meter to make measurements. I use either a Mitey Mike II, an EarthWorks QTC40 or a DPA 4006 (all calibrated to be flat on-axis), with either an AudioControl SA3050A spectrum analyzer or sotware-based analysers.

John Atkinson
Editor, Stereophile

  • X
    Enter your Stereophile.com username.
    Enter the password that accompanies your username.
    Loading