My speakers in room bass response

[3db]

I had generated some test tones and burned them on a CD running the full gambit from 20Hz ro 20KHz . I was curious to see my speakers bass response and room interaction so here are my findings.

From my seating position, using a 1KHz test tone, I chose a reference value of 74db at the seated locatin. Holding the Rat Shack meter within 6" of the ceiling, I measured 77db.

Frequency; Seated Postion (db); Ceiling (db)
100; 81; 81
90; 82; 81-82
80; 79; 79
70; 76; 78
60; 70; 75
50; 73; 75
40; 61; 64
30; 73; 74
20; 52; 52

The dip at 40 surpised me a little. I also have a rather nast y echo if I clap my hands in that room. Will acoustic treatments smooth out the bass response, particluairy at 40 Hz and will it remove the echo that I expereience when clapping my hands?

[dmb_fan]

Yeah, it's a kick in the pants when one finally measure their LF response with some test tones. Rest assured, even with a 20dB difference between 90Hz and 40Hz, most folk's rooms are no better. Keep in mind that these numbers don't tell you much about how music sounds in the room. For a better idea, you need to do the tests with 1/6th octave test tones and for the best idea of what's going on, you'll need to use a full spectrum frequency sweep.

Also keep in mind that in this scenario, you don't know if 74dB is "flat" or not. You'll want to do the entire frequency spectrum (or a critical part, say 500Hz to 1Khz) at 1/6th octave to find out what the average level is. Your speakers might be a little up or a little down at 1Khz, so without contextual measurements, you don't know what level your bass response should be.

There are very few room treatments that will help with that 40Hz null. The Venus bass trap from auralex.com is rated to absorb frequencies that low, but it probably won't do much at 40Hz. They also have a Mega LENRD that is rated down to 50Hz, but again, at the 50Hz level it probably won't do a whole lot to smooth the frequency response.

The good news is that very little musical information actually happens at 40Hz. 41Hz is the lowest note on a bass guitar, and 82Hz is the lowest note on a six-string guitar.

Also, I'm curious about that 73dB at 30Hz. Does the tone you hear actually sound lower than the tones you hear at 40 or 50Hz? I ask because when I tested my speakers (which are rated by the manufacturer flat to 20Hz) the tone I hear at 20Hz is really just harmonic distortion. A subwoofer playing 20Hz cleanly is all but inaudible to me. My speakers playing 20Hz is VERY audible. I'm just pointing out that there is probably some significant distortion in that 30Hz number, unless of course you've got some pretty beefy speakers.

There are a couple things you can do to improve the flatness of this response:
1. Speaker position. This link has some helpful suggestions for speaker positioning that might help to reduce the impact of room nodes.
http://www.cardas.com/content.php?area=insights&content_id=26&pagestring=Room+Setup

2. Room treatments. Like I said, the Venus and the LENRD are only good down to 40 or 50Hz, but there's very little musical information below these notes. And yes, the addition of foam to your room will help with slap-echo.

3. Dedicated subwoofer. A dedicated sub placed in a corner and handling everything below 80Hz is likely to produce a flatter and stronger bass response than two speakers placed at the optimal locations for imaging and high frequency response. You can also then insert a parametric EQ to flatten the sub's response at the listening position.

This SynAudioCon test topic on bass traps should give you an idea of what is realistically possible with foam room treatments. It's the bottom link on this page:
http://auralex.com/press/reviews.asp

Anyhow, I hope that helps. And I hope some of the guys on this forum who REALLY know about this stuff pitch in here. I'm just a casual observer, but no one had answered your post so I thought I'd try to help.

-Adam

[dmb_fan]

Here are a couple graphs of in-room response I've done from my listening position. I thought they might interest you, if for nothing else because they are labelled with the correct 1/6th octave test tones you'll want to use if you want to pursue this further.

For what it's worth, the response from the Thiels is far less flat then that from the Hsu/Snell combination, but the Thiels sound infinitely better with both music and movies. Go figure.

Hope it helps!

-Adam

[3db]

Yeah, it's a kick in the pants when one finally measure their LF response with some test tones. Rest assured, even with a 20dB difference between 90Hz and 40Hz, most folk's rooms are no better. Keep in mind that these numbers don't tell you much about how music sounds in the room. For a better idea, you need to do the tests with 1/6th octave test tones and for the best idea of what's going on, you'll need to use a full spectrum frequency sweep.


Also keep in mind that in this scenario, you don't know if 74dB is "flat" or not. You'll want to do the entire frequency spectrum (or a critical part, say 500Hz to 1Khz) at 1/6th octave to find out what the average level is. Your speakers might be a little up or a little down at 1Khz, so without contextual measurements, you don't know what level your bass response should be.
-Adam

Sounds like a 2 week exercise if doen manually. I just wonder how aacurate it will be. As to your comment about 74db being flat.. maybe I should look at the anoechic chamber frequ plots for my speakers (PSB Image T45s) and pick a frequency that is close to the zero db mark as possible and use that as my ref frequency. All in all though, I'm still dealing with room interaction I think,,maybe no tso much at teh 1Khz range


There are very few room treatments that will help with that 40Hz null. The Venus bass trap from auralex.com is rated to absorb frequencies that low, but it probably won't do much at 40Hz. They also have a Mega LENRD that is rated down to 50Hz, but again, at the 50Hz level it probably won't do a whole lot to smooth the frequency response.

The good news is that very little musical information actually happens at 40Hz. 41Hz is the lowest note on a bass guitar, and 82Hz is the lowest note on a six-string guitar.
-Adam

Could I reduce that 40Hz dip my moving the speakers around?


Also, I'm curious about that 73dB at 30Hz. Does the tone you hear actually sound lower than the tones you hear at 40 or 50Hz? I ask because when I tested my speakers (which are rated by the manufacturer flat to 20Hz) the tone I hear at 20Hz is really just harmonic distortion. A subwoofer playing 20Hz cleanly is all but inaudible to me. My speakers playing 20Hz is VERY audible. I'm just pointing out that there is probably some significant distortion in that 30Hz number, unless of course you've got some pretty beefy speakers.
-Adam

Yup, I definately noticed it play lower and louder at 30 Hz than I did at 40 Hz . The loudness was perceived to be the same for 50 60 Hz

Like to thank you for your help

[dmb_fan]

Could I reduce that 40Hz dip my moving the speakers around?


At that frequency, it's probably more about room geometry than speaker location, but it's worth a shot. Play a 40Hz tone on repeat at a reasonable volume and move the speakers around.

If the dip is less than 1/6th octave or so, then you have nothing to worry about. It won't be very noticeable at all except with sine waves.

Have you ever watched a spectrum analyzer? They're often on inexpensive receivers. The Technics I had in high school had one. Think about what happens when a bass note (or any note) is played. It's not like the one bar at 40Hz rises up. Three or four bars on the left side of the analyzer will rise up. You've got the fundamental and the harmonics and all the little nuances that make a Fender bass sound different from an Ibanez bass. If your null at 40Hz is less than 1/6th octave, all of these things will play a part and you'll never be able to hear the null.

About the 74dB thing... it won't take long to find zero. Just do the 1/6th octave tones between say 500Hz and 1KHz. Six tones. Find the average, and that's your new zero mark.

This CD might help:
http://basszone.stryke.com/testcd.html
It's got 1/6th octave tones for like the first 3.5 octaves (20Hz to 200Hz or so) and then fewer tones per octave.

I saw a post on fiddling with subwoofer settings. Are these the results from just your main speakers or from your main speakers with the subwoofer. Perhaps the 40Hz null is caused be out of phase info from speakers and sub cancelling each other out.

One thing I've thought about doing is sending my frequency response graph to Auralex with a sketch of my room and saying, "What should I do to fix this?" I hear that they're incredible with customer service and some of the staff really will latch onto a problem and try to fix it. Like, they'll bug you by email to ask if their solution worked. Pretty cool.

:)

-Adam

[3db]

At that frequency, it's probably more about room geometry than speaker location, but it's worth a shot. Play a 40Hz tone on repeat at a reasonable volume and move the speakers around.

If the dip is less than 1/6th octave or so, then you have nothing to worry about. It won't be very noticeable at all except with sine waves.

Have you ever watched a spectrum analyzer? They're often on inexpensive receivers. The Technics I had in high school had one. Think about what happens when a bass note (or any note) is played. It's not like the one bar at 40Hz rises up. Three or four bars on the left side of the analyzer will rise up. You've got the fundamental and the harmonics and all the little nuances that make a Fender bass sound different from an Ibanez bass. If your null at 40Hz is less than 1/6th octave, all of these things will play a part and you'll never be able to hear the null.
-Adam

I understand where your coming from comong an electrical engineering background. I'm not familiar with the human physiology side when it comes to octaves. I do understand how spectal content will alter the sound like the difference of the two guitars you mentioned in the example.


About the 74dB thing... it won't take long to find zero. Just do the 1/6th octave tones between say 500Hz and 1KHz. Six tones. Find the average, and that's your new zero mark.

This CD might help:
http://basszone.stryke.com/testcd.html
It's got 1/6th octave tones for like the first 3.5 octaves (20Hz to 200Hz or so) and then fewer tones per octave.

I saw a post on fiddling with subwoofer settings. Are these the results from just your main speakers or from your main speakers with the subwoofer. Perhaps the 40Hz null is caused be out of phase info from speakers and sub cancelling each other out.

One thing I've thought about doing is sending my frequency response graph to Auralex with a sketch of my room and saying, "What should I do to fix this?" I hear that they're incredible with customer service and some of the staff really will latch onto a problem and try to fix it. Like, they'll bug you by email to ask if their solution worked. Pretty cool.

:)

-Adam

Thanks for your help. My measurements were strictly using the mains only , no sub. The speakers are PSB Image T45s.

One thing I did think off . I generated the test tones using some freeware software. I left the setting alone that controlled the level of the test tones generated . I wonderif the signal strength generated deviated from one frequency to the next.