Wooch et al... My First SPL Readings [Archive] - Audio & Video Forums

PDA

View Full Version : Wooch et al... My First SPL Readings



the hand of boredom
06-06-2004, 07:28 PM
My mummy had sent me some cash for my birthday last week. Took a trip down to the local Rat Shack and picked me up one of 'em SPL devices. I came home, downloaded since waves ranging from 14 Hz to 120 Hz from snapbug.ws.

I have the mains (Studio/40v2) set to LARGE on my SR 4300. The LFE is sent to the X30 crossover and split to the two Servo 15s. The crossover is set very near 70 Hz. The listening room is set up in a basement with carpet over a concrete floor (no sub-floor) with three exterior walls (front, left wall, and back wall). The ceiling is 8 ft, with acoustic ceiling panels. Front-to-back wall distance is 18ft, the room is 14ft across the back half, and 16 ft across the front, with a walk-way that is open to the rest of the house.

Now I'm no expert, but I think that something is drasticly wrong with my audio environment. As the sines go from 20 to 35 Hz, the SPL increase sharply from 84db to 99db. The levels dip steadily to 68db as we near 70 Hz, with a peak centred around 55 Hz reaching 81db. Once past 70 Hz, the levels rise to 82db near 80 Hz and hover in the mid 80s to 90 Hz. Between 90 and 120 Hz, the levels rise and fall between 70 and 80 Hz.

Are the lowest frequencies normally supposed to be much LOWER than the rest of the spectrum? Yes, I had C weighting. Yes, double checked everything. I first thought that maybe the subs are out-of-phase with each other, but then there would be 0db at the lowest frequencies. I thought that maybe because the room is open to the house, that some pressure would be lost, but the 20-30 hz jump still wouldn't make sense. The only explanation that I can think of is that the subs are out of phase with the mains (?).

I still need to do the following:

1. run the sines through the mains with the subs turned off
2. run the sines through the subs only with the mains turned off
3. run the sines through each sub separately to see how placement affects response.

I did get a chance to make on breath-taking (and viscera-shaking measurement). I ran the 20 hz wave a couple times with the reader one meter from both subs and turned to volume up a bit until it just didn't seem right. I was able to achieve levels in extreme excess of 126 db, the needle was stuck in the red zone.

Anyway, can anyone make sense of the numbers I gave above? I will get some graphs going to make it easier to see what is going on. I will also run the tests above and keep you all updated with the findings.

THOB

mtrycraft
06-06-2004, 07:53 PM
My mummy had sent me some cash for my birthday last week. Took a trip down to the local Rat Shack and picked me up one of 'em SPL devices. I came home, downloaded since waves ranging from 14 Hz to 120 Hz from snapbug.ws.

I have the mains (Studio/40v2) set to LARGE on my SR 4300. The LFE is sent to the X30 crossover and split to the two Servo 15s. The crossover is set very near 70 Hz. The listening room is set up in a basement with carpet over a concrete floor (no sub-floor) with three exterior walls (front, left wall, and back wall). The ceiling is 8 ft, with acoustic ceiling panels. Front-to-back wall distance is 18ft, the room is 14ft across the back half, and 16 ft across the front, with a walk-way that is open to the rest of the house.

Now I'm no expert, but I think that something is drasticly wrong with my audio environment. As the sines go from 20 to 35 Hz, the SPL increase sharply from 84db to 99db. The levels dip steadily to 68db as we near 70 Hz, with a peak centred around 55 Hz reaching 81db. Once past 70 Hz, the levels rise to 82db near 80 Hz and hover in the mid 80s to 90 Hz. Between 90 and 120 Hz, the levels rise and fall between 70 and 80 Hz.

Are the lowest frequencies normally supposed to be much LOWER than the rest of the spectrum? Yes, I had C weighting. Yes, double checked everything. I first thought that maybe the subs are out-of-phase with each other, but then there would be 0db at the lowest frequencies. I thought that maybe because the room is open to the house, that some pressure would be lost, but the 20-30 hz jump still wouldn't make sense. The only explanation that I can think of is that the subs are out of phase with the mains (?).

I still need to do the following:

1. run the sines through the mains with the subs turned off
2. run the sines through the subs only with the mains turned off
3. run the sines through each sub separately to see how placement affects response.

I did get a chance to make on breath-taking (and viscera-shaking measurement). I ran the 20 hz wave a couple times with the reader one meter from both subs and turned to volume up a bit until it just didn't seem right. I was able to achieve levels in extreme excess of 126 db, the needle was stuck in the red zone.

Anyway, can anyone make sense of the numbers I gave above? I will get some graphs going to make it easier to see what is going on. I will also run the tests above and keep you all updated with the findings.

THOB

For starters, you need to know if th esignals you downloaded were the same level? Second, there is a correction factor to the SPL meter. You need to add these to the numbers:

16 Hz +11.5;
20 Hz + 7.5;
25 Hz +5;
31.5 Hz +3;
40 Hz +2.5;
50 Hz +1.5;
63 Hz +1.5;
80 Hz +1.5;
100 Hz +2.
Then your curve doesn't look so bad:)

Woochifer
06-07-2004, 11:42 AM
I don't see anything abnormally wrong with your readings. That extreme variation at specific frequencies is pretty typical for a medium sized room like yours. The dip as you go from 70 Hz to 80 Hz seems like something you can remedy by adjusting the crossover settings (possibly going to a lower crossover frequency), however this was created by a room-induced null, then your only alternative is to try different placements or move your listening position. Keep in mind that the Studio 40s and Servo 15s will have different low frequency issues because they are placed in different locations.

Also if you're using the X-30 with the Servo 15, make sure that any crossover settings on the Servo 15 itself is bypassed (or if bypass unavailable, turn the crossover setting all the way up on the Servo 15 since the X-30 handles the bass management).

Given how huge the peak is around 30 Hz, you might need to go with a parametric EQ if different placements can't remedy the problem. On my in-room subwoofer measurement, you can see how the parametric EQ evened out the frequency response (this chart uses the correction values that mtry mentioned). I had problems at 32 and 88 Hz (crossover frequency on my receiver is fixed at 90 Hz).

http://hometown.aol.com/sfwooch/myhomepage/subtest.gif

the hand of boredom
06-07-2004, 08:27 PM
For starters, you need to know if th esignals you downloaded were the same level? Second, there is a correction factor to the SPL meter. You need to add these to the numbers:

16 Hz +11.5;
20 Hz + 7.5;
25 Hz +5;
31.5 Hz +3;
40 Hz +2.5;
50 Hz +1.5;
63 Hz +1.5;
80 Hz +1.5;
100 Hz +2.
Then your curve doesn't look so bad:)

Mtry:

I would think that the sines were all of the same frequency, the input meter on the winamp display appears to be the same all the way through.

What's the deal with the correction levels? The manual did not say anything about it.

THOB

the hand of boredom
06-07-2004, 08:32 PM
Keep in mind that the Studio 40s and Servo 15s will have different low frequency issues because they are placed in different locations.



Wooch Man:

I stand you corrected. I have the 40s placed directly on top of the 15s with lots of blu tak between'em.

Now I remember back in the day of R. Greene he complained about a 16db peak at a particular frequency which he corrected to a 6db difference. Looking at my readings, there is a difference of almost 30db at certain points. This can't be normal can it?

This laptop needed to be formatted today, the graphs and all my numbers have been erased. So I need to run this all again. It will be a day or two before I can get cracking again.


THOB

mtrycraft
06-07-2004, 09:32 PM
Mtry:

I would think that the sines were all of the same frequency, the input meter on the winamp display appears to be the same all the way through.

What's the deal with the correction levels? The manual did not say anything about it.

THOB


The RS meter manual does show how it is not linear over its frequency band.
Some have measured it and developed this table. If your levels were the same across the band, your signals would read low by the amount I posted. You have to add that number to know the real level the speaker is off die to the room factors.

Richard Greene
06-09-2004, 04:04 PM
The typical room measured at the listening position will be worse than +/-10dB and typically +/-15dB even with the flattest speakers money can buy. Subwoofers generally have a flat frequency response when measured near field (one inch from dust cap) although the microphone's proximity effect artificially boosts the measurements a few dB's in the 50-100Hz. range.

Your numbers reminded me that during some experiments last week moving all my speakers closer to my ears, moving my subwoofer a mere two feet closer created the worst measurements yet in my room (from a null of 68dB to a peak of 96dB measured at
1/6 octave intervals (would have been a few dB worse measured with higher resolution).

For the first time in over 20 years I was unable to figure out EQ settings to fix the problems within one hour and then gave up.

But after I lifted up my tube sub to the vertical position it was originally designed for
(firing up with the driver four feet off the ground), the frequency response improved to the best I have ever measured in my room. Still three bands of parametric EQ were required
(probably about three 1/6 octave -6dB cuts) which was accomplished in about 10-15 minutes!

Funny thing about bass inside a room is that it varies tremendously when you move your ears only a foot or two or move the sub four or five feet.

mtrycraft
06-09-2004, 09:20 PM
Where have you been?

Richard Greene
06-10-2004, 07:05 AM
Not that any would admit even the possibility they could not hear a difference between two components ... because they know what they hear ... and their ears are more accurate than any meters ... just don't ask them to prove it !.

Sir Terrence the Terrible
06-10-2004, 08:01 AM
Not that any would admit even the possibility they could not hear a difference between two components ... because they know what they hear ... and their ears are more accurate than any meters ... just don't ask them to prove it !.

Leave the inmates alone, we need you here!!! I have had a difficult time explaining your absence, and have to use such lame(and untrue) excuses like " he is actually sober and hard at work". Everyone knows this couldn't be true

mtrycraft
06-10-2004, 09:50 PM
Not that any would admit even the possibility they could not hear a difference between two components ... because they know what they hear ... and their ears are more accurate than any meters ... just don't ask them to prove it !.


I guess I missed you there then. I read jneutron and Steve Eddy at prop.

Smokey
06-11-2004, 11:41 AM
I have had a difficult time explaining your absence, and have to use such lame(and untrue) excuses like " he is actually sober and hard at work".

Yes, he is sober and hard at work in the Asylum, creating chaos :D