Understanding & Using Test Tones To Calibrate Your System

[Big Daddy]

For additional information, read Calibrating Your Audio With an SPL Meter and Calibrating Your Subwoofer(s) with the Velodyne SMS-1 Equalizer/Analyser.

INTRODUCTION

All modern receivers and preamp/processors include Pink Noise and a program such as Audyssey for automatic calibration of your audio system. Unfortunately, because of the anomalies that are created with multiple speakers and subwoofers in a small home theater room, an external 1/3 octave equalizer/analyzers may have become a necessity to tame the interaction of waves and room boundaries. Furthermore, when you use the pink noise from your receiver to adjust the subwoofer level, you will not get accurate results. The main reason is that your calibration program or the SPL meter (if you are using one) measure the peak frequency generated by the sub at that listening position and all other frequencies are obscured. This is why it is best to use test tones and an SPL meter for a subwoofer calibration.

Because of the reasons discussed above, a thorough understanding of Test Tones, Octaves, 1/3 Octaves, Pink Noise, White Noise, and Brown Noise are required.

OCTAVES AND 1/3 OCTAVES

Definition of Octave
Assume one note has a frequency of 200Hz, the note an octave above it is at 400Hz, and the note an octave below is at 100Hz. The ratio of frequencies of two octaves is 2:1. Another example: 50Hz is one octave below 100Hz, and 400Hz is two octaves above 100Hz.

An Octave band is characterized by the following formulas:

If the F1 is the lower cutoff frequency and F2 is the upper cutoff frequency, the ratio of the band limits is given by:

F2/F1 ≈ 2^N

where N = 1 for full Octave and N= 1/3 for 1/3 Octave bands.

Amplitude Versus Frequency

An Octave has a center frequency that is 2^(1/2) (square root of 2) times the lower frequency cutoff frequency and has an upper cutoff frequency that is twice the lower cutoff frequency. Therefore,

F1 = F0 / 2^(1/2)
F2 = 2^(1/2) . F0, where F0 is the center frequency
F2 = 2 . F1
BW = F2 - F1, where BW = Band Width

A 1/3 Octave has a center frequency that is 2^(1/3) (cubic root of 2) times the lower frequency cutoff frequency and has an upper cutoff frequency that is twice the lower cutoff frequency.

Therefore,

F1 = F0 / 2^(1/3)
F2 = 2^(1/3) . F0

Example 1. Calculate the 1/3 Octave band limits for center frequency of 16Hz.

2^(1/3) = 1.26

Lower band limit: F1 = 16Hz / 1.26 ≈ 12.5Hz
Upper band limit: F2 = 16Hz * 1.26 ≈ 20Hz

Example 2. Calculate the 1/3 Octave band limits for center frequency of 125Hz.

2^(1/3) = 1.26

Lower band limit: F1 = 125Hz / 1.26 ≈ 100Hz
Upper band limit: F2 = 125Hz * 1.26 ≈ 160Hz

Octave Bands



One-Third Octave Bands




PINK NOISE, WHITE NOISE, AND BROWN NOISE

White Noise
White noise has the same distribution of power for all frequencies, so there is the same amount of power between 0 and 500Hz, 500Hz and 1,000Hz or 20,000Hz and 20,500Hz. This means that the band level increases by 3dB with every Octave. In other words, if you double the frequency, the sound pressure level will increase by 3 dB. For example, assume that we move from the first octave (20Hz to 40Hz) to the next octave (40Hz to 80Hz). The white noise has the same level of power for the 20Hz to 40Hz, 40Hz to60Hz and 60Hz to 80Hz. This means that the second octave has twice as much power as the first octave. The formula for dB level is defined as 10 . log(Power2/Power1) and since Power2= 2 Power1, the formula is simplified to 10. log (2) or 3 dB.

White noise is normally the random noise that we hear as hiss created by all electronic circuits. White noise is also created in nature as wind blowing through leaves or on the beach. White noise has equal energy for each frequency width (cycle). As the frequency (cycles) doubles for each octave, so does the noise energy (+3dB/octave), resulting in white noise sounding as treble sound or hiss.

Pink Noise
Pink noise has the same distribution of power for each octave, so the power between 0.5Hz and 1Hz is the same as between 5,000Hz and 10,000Hz. It has a linear scale so that for each octave band, it will produce the same amplitude. Pink noise is filtered white noise so each octave has equal energy, therefore a flat energy response and is similar to music and useful for acoustic measurements for sound system calibration. Remember that sound system calibration by Pink noise is a simplified approximation only and is not an accurate method.

White Noise Versus Pink Noise Plot
The energy distribution of pink noise is flat, octave-wise. On a logarithmic frequency plot, pink noise is represented by a flat horizontal line. Pink noise sounds evenly spread across all frequencies and best approximates the spectral distribution of music and the human auditory system, which processes frequencies logarithmically.

The energy distribution of white noise is flat which means that it contains all frequencies in equal proportion. Because the human auditory system processes audio in a logarithmic frequency scale, white noise sounds much brighter than pink noise.

Brown Noise
Brown noise has more energy at lower frequencies, even more than pink noise. Brown noise produces a warmer tone compared to white noise or pink noise. Brown noise is useful to check subwoofer phase issues. It can also be used for burning in midrange and bass drivers. To properly integrate your subwoofer into your audio system, the subwoofer should be in phase with the other drivers at the crossover point.

How to Adjust Your Subwoofer
As I indicated before, using pink noise from your receiver to adjust the subwoofer level will not give you accurate results. In addition, when you use pink noise with an SPL meter, you will notice that the meter is not very stable and its reading varies quite a bit. That is quite normal. Even if you move a couple of inches, the reading may change. It is the nature of long wavelength bass frequencies and their interaction with the room boundaries. It is for this reason that the use of test tones and an SPL meter are preferred.

Using a test tone disc, adjust the master volume on the receiver so that the SPL meter reads 75dB with a 50Hz tone at the listening position. Take measurements of four different tones 1/3 octave above the 50Hz and four different tones 1/3 octave below 50Hz. Average together each set of four measurements and adjust the subwoofer’s gain level to match the other speakers. This is explained in the following:

1. First of all, adjust the levels of all the speakers using the internal test tones of the receiver so that the SPL meter reads 75dB for all the speakers.
2. Download pink noise and 1/3 octave test tones and copy them to a disc. Alternatively, you can purchase a test tone calibration disc.
3. Run the pink noise from the external disc through the speakers and adjust the master volume on the receiver so that the SPL readings from your front, center, and surround speakers are 75dB.
4. Do not touch the master volume after that.
5. Set the subwoofer's gain in the receiver's menu to zero and play the 50Hz test tone.
6. Adjust the level on the back of the subwoofer until the SPL reading is 75dB.
7. After this, do not touch the level on the back of the sub.
8. Play 4 test tones 1/3 octave below 50Hz and 4 test tones 1/3 octave above 50Hz and write down their SPL numbers on a piece of paper or enter them in a spreadsheet such as Excel.
9. The test tones 1/3 octave apart from 50Hz are:
   20Hz, 25Hz, 31.5Hz, 40Hz, 50Hz, 63Hz, 80Hz, 100Hz, 125Hz.
10. Average the 9 numbers to get 75dB. In excel, it is very easy. Assuming the nine numbers are in cells A1 through A9, the function =Average(A1:A9) does that automatically.
11. If the average is above 75dB, turn the subwoofer gain in the receiver's menu down and take the measurements again. If the average is below 75dB, increase the subwoofer gain in the receiver's menu and repeat the experiment.
12. Your job is done when the average of the 9 numbers is 75dB.
13. If you like more bass, it is ok to make the average of the 9 test tones slightly higher than 75dB.

In case you want to use 1/6 octave apart frequencies to calibrate your subwoofer, these are the frequencies:
16Hz, 18Hz, 20Hz, 22Hz, 25Hz, 28Hz, 31.5Hz, 36Hz, 40Hz, 45Hz, 50Hz, 56Hz, 63Hz, 71Hz, 80Hz, 89Hz, 100Hz, 111Hz, 125Hz, 142.5Hz, 160Hz.

Subwoofer Calibration with Rives Audio Test-CD
Subwoofer Calibration Using Rives Audio Test CD 2 — Reviews and News from Audioholics

Excel Worksheet
It is highly recommended to run 1/3 octave frequencies through the subwoofer and plot the output level in dB versus versus frequency by hand or using the Microsoft Excel program. Alternatively, you can download a free Excel Worksheet. Here is Another Excel Worksheet to adjust your subwoofer.

Low Frequency Response and Subwoofer Test
Play back the tone until you start hearing the tone as it rises. This frequency more or less represents the lowest limit of your audio system.
Low Frequency Response and Subwoofer Sound Test

High Frequency Response and Hearing Test
Play back the tone until you start hearing the tone as the frequency drops. This frequency more or less represents the upper limit of your audio system, or your hearing.
High Frequency Response and Hearing Test

Audible Dynamic Range Test
Play the tone until you can't hear the voiceover anymore. The dynamic range of system is approximately given by the level the voice message was playing at when it was still (barely) audible. The limiting factor here will be the environmental noise in your listening room (approximately 40dBA).
Audible Dynamic Range Test

Left/Right Stereo Test
This test tone is for the checking the connection of the right and left speakers.
Left/Right Stereo Sound Test

Stereo Polarity Test
Depending on how a speaker has been wired internally or connected to the amplifier, it will either move in or out in response to a given input signal.
Stereo Polarity Sound Test

Pink Noise
Pink noise is used for calibration and to check the frequency responses or the adverse effects of room modes. It can also be used for burning in speaker drivers.
Pink Noise

White Noise
White noise is used to check frequency responses or the adverse effects of room modes.
White Noise

Brown Noise
Brown noise is useful to set a subwoofer's phase with respect to the other speakers. It can also be used for burning in midrange and bass drivers.
Brown Noise

Swept Sine (20Hz to 20kHz)
These test tones are used to check frequency response or the adverse effects of room modes.
Swept Sine (Full Spectrum)

Swept Sine (20Hz to 200Hz)
Use these test tones to determine resonant frequencies in your room.
Swept Sine (Lower Spectrum)

[Big Daddy]

Free Test Tones:
http://www.snapbug.ws/sinewaves/
http://www.stompaudio.com/test-tones-c-6793.html
http://www.realmofexcursion.com/downloads.htm
http://audiocheck.net/testtones_index.php
http://www.realtraps.com/test-cd.htm
http://www.cleansofts.com/softdownlo...Generator.html
http://www.nch.com.au/tonegen/index_b.html

Sound Doctor ($10):
http://www.soundoctor.com/testcd/
Rives Audio Test Tones ($21):
http://www.rivesaudio.com/order/orderframe.html
kja Scientific Audio Resources ($19.99):
http://www.rainfall.com/cdroms/pink_instructions.htm
Stereophile 3 Test CD's ($24.95):
http://ssl.blueearth.net/primedia/pr...age=1&featured
My Disc: Autosound 2000 Test Disc:
http://www.amazon.com/My-Disc-Autoso...9721453&sr=1-2
Many Test CD's on Amazon:
http://www.amazon.com/gp/bestsellers...rsr_m_1_3_last

Subwoofer Test Clips:
http://www.dragtimes.com/video-viewer.php?v=PC7WEQsiz08

crazy subwoofer test tones - YouTube

YouTube

NEW BASS TESTS - YouTube

YouTube

Lil Jon - Bass Test - YouTube

BASS TEST - YouTube

http://www.metacafe.com/tags/subwoofer/most_popular/
http://www.stealthsettings.com/woofe...fer-sound-test

Free Hearing Test:
http://lahave-audio.com/

ADDITIONAL INFORMATION AND REFERENCES

http://zone.ni.com/devzone/cda/tut/p/id/101
http://en.wikipedia.org/wiki/Octave
http://www.cross-spectrum.com/audio/...equencies.html
http://www.zytrax.com/tech/audio/calculator.html
http://www.zytrax.com/tech/audio/equalization.html
http://eceserv0.ece.wisc.edu/~sethares/consemi.html
http://www.sengpielaudio.com/calculator-octave.htm
http://en.wikipedia.org/wiki/White_noise
http://en.wikipedia.org/wiki/Pink_noise
http://www.bigpicturebigsound.com/article_560.shtml
http://www.harman.com/wp/pdf/Loudspe...ndRoomsPt1.pdf
http://www.harman.com/wp/pdf/Loudspe...ndRoomsPt2.pdf
http://www.harman.com/wp/pdf/Loudspe...ndRoomsPt3.pdf

[Driver_King]

BD is at it again! It looks good so far.

[Big Daddy]

Originally Posted by Driver_King

BD is at it again! It looks good so far.

BD has completely lost his mind.

Blame it on Aramis. He asked the question in the Upgrading Subwoofers thread.

[rded]

I didn't know class is in session on a Saturday evening? I've got sooooo much to learn.

[Big Daddy]

Originally Posted by rded

I didn't know class is in session on a Saturday evening? I've got sooooo much to learn.

People ask simple questions without realizing that there are no simple answers.

[rded]

Originally Posted by Big Daddy

People ask simple questions without realizing that there are no simple answers.

As usual we appreciate it! I just too much info for us simple minded folks to digest

[crazyBLUE]

Originally Posted by rded

As usual we appreciate it! I just too much info for us simple minded folks to digest

I'll second that !

As again confused as he-- ! So this will be A 2 ~ 3 ~ & maybe 4th read for me . Thats just to have an idea of what is happening ! There was the word calibration in there & today I moved my sub sum & recalibrated but changed A few things before A movie & I don't think I need A sub anymore ! My Fronts took care of it !

[jomari]

darn it! yet another sticky! man, i gotta step up my game then...

[aramis109]

Originally Posted by Big Daddy

People ask simple questions without realizing that there are no simple answers.

I think I'm more confused now. LOL... if my crossover frequency was 80hz, what would be the testing points above and below it?

It's ok to cheat and just give me the answer on this one- I swear I'll never figure it out on my own.

[zicmubleu]

Originally Posted by Big Daddy

BD has completely lost his mind.

Blame it on Aramis. He asked the question in the Upgrading Subwoofers thread.

This is some interresting information, once you understand why you want it.

I think it would help to show why you created this post, even just a reference to aramis109's original question about using his spl meter and what frequencies to measure - I think that is what is being discussed here. I found post 464 in your upgrading subwoofer which I believe was the spark for this thread.

It might be useful to add the scale on the graph showing how the frequncy is going up in a doubling fashion, but maybe it would get too busy. However looking at the chart gives the initial impression that the center frequency is actual the mathematical center frequency because it looks physically centered on the diagram.

[zicmubleu]

Originally Posted by aramis109

I think I'm more confused now. LOL... if my crossover frequency was 80hz, what would be the testing points above and below it?

It's ok to cheat and just give me the answer on this one- I swear I'll never figure it out on my own.

I am guessing it will be 63 and 100 HZ based on the chart, but I am not clear on how to apply this. Using headphones until I get some speakers.

[aramis109]

Originally Posted by zicmubleu

This is some interresting information, once you understand why you want it.

I think it would help to show why you created this post, even just a reference to aramis109's original question about using his spl meter and what frequencies to measure - I think that is what is being discussed here. I found post 464 in your upgrading subwoofer which I believe was the spark for this thread.

It might be useful to add the scale on the graph showing how the frequncy is going up in a doubling fashion, but maybe it would get too busy. However looking at the chart gives the initial impression that the center frequency is actual the mathematical center frequency because it looks physically centered on the diagram.

You got it right. I agree, it'd maybe be helpful to add the "why" to the OP or the 2nd post.

Originally Posted by zicmubleu

I am guessing it will be 63 and 100 HZ based on the chart, but I am not clear on how to apply this. Using headphones until I get some speakers.

As far as application, I have a few options for playing test tones. Using my SPL meter I can make sure the sub is hitting the appropriate 75db so that it's dialed in to the rest of the system. For the short term, I just adjusted the gain a bit since moving it onto an Auralex Gramma. Haven't had time to calibrate what with my wife on full-on nesting mode.

[Big Daddy]

Originally Posted by aramis109

I think I'm more confused now. LOL... if my crossover frequency was 80hz, what would be the testing points above and below it?

It's ok to cheat and just give me the answer on this one- I swear I'll never figure it out on my own.

For adjusting the level of your subwoofer with an SPL meter if your crossover frequency is 80Hz, then the 4 frequencies 1/3 octave below it are 31.5Hz, 40Hz, 50Hz, 63Hz and the 4 frequencies 1/3 octave above it are 100Hz, 125Hz, 160Hz, 200Hz.

[Big Daddy]

Originally Posted by zicmubleu

This is some interresting information, once you understand why you want it.

I think it would help to show why you created this post, even just a reference to aramis109's original question about using his spl meter and what frequencies to measure - I think that is what is being discussed here. I found post 464 in your upgrading subwoofer which I believe was the spark for this thread.

It might be useful to add the scale on the graph showing how the frequncy is going up in a doubling fashion, but maybe it would get too busy. However looking at the chart gives the initial impression that the center frequency is actual the mathematical center frequency because it looks physically centered on the diagram.

I added significant amount of information to the original post. I hope it is more clear now. I will add more information in the future. As I said it in the OP, I am only half human.

[aramis109]

Originally Posted by Big Daddy

For adjusting the level of your subwoofer with an SPL meter if your crossover frequency is 80Hz, then the 4 frequencies 1/3 octave below it are 31.5Hz, 40Hz, 50Hz, 63Hz and the 4 frequencies 1/3 octave above it are 100Hz, 125Hz, 160Hz, 200Hz.

Thanks. I really was lost on this one. I did manage to figure it out thanks to Hsu's excel spreadsheet cheat sheet that I found in another thread, and it matches up exactly with what you posted (not that I would doubt you, just adding in secondary confirmation).

I probably won't have the chance to do this for awhile since my free time has been filled with my "honey do" list as my wife is in full nesting mode for the baby. After things settle down I'll definitely give this a run however. I also need to re-run Audyssey since I added an isolation riser and run some more speaker wire to my surround since I have tweaked their placement a bit and the wire is a little taut.

[liquidice]

so today I had a friend hook me up with a couple of personal custom test-tone cd's for calibrating my subwoofer's. I'll be getting my SPL meter tomorrow. With two subs, I would assume I would obviously calibrate each one seperately. But after I do that, and I turn both on, and the bass is now louder than the other 5 speakers, would I tweak further?...and actually is that even possible? Better example, if one sub's gain is set to output about 70db, and then I do the same with the other sub, then run both at same time, would output still be 70db? Anything different I would do having two subs? Also, if BD or aramis could link the HSU excel sheet, I'm curious what it is. Each of my 5 speakers have different xover points, so this could be interesting.

[jomari]

Originally Posted by liquidice

so today I had a friend hook me up with a couple of personal custom test-tone cd's for calibrating my subwoofer's. I'll be getting my SPL meter tomorrow. With two subs, I would assume I would obviously calibrate each one seperately. But after I do that, and I turn both on, and the bass is now louder than the other 5 speakers, would I tweak further?...and actually is that even possible?

Calibrate the first/primary subwoofer first using the guide, then turn it off.

Calibrate the secondary subwoofer, following the same guidelines, then turn it off.

Calibrate the other speakers - which would be so much easier compared to calibrating the subs...

Better example, if one sub's gain is set to output about 70db, and then I do the same with the other sub, then run both at same time, would output still be 70db? Anything different I would do having two subs?

You SHOULD ideally have the output set at 70db, but with a better, even response compared to a single subwoofer.

Also, if BD or aramis could link the HSU excel sheet, I'm curious what it is. Each of my 5 speakers have different xover points, so this could be interesting.

harman white papers

all articles are definitely worth your time, taking a second read on most of em to finally get it.

your welcome.

[Big Daddy]

Originally Posted by liquidice

so today I had a friend hook me up with a couple of personal custom test-tone cd's for calibrating my subwoofer's. I'll be getting my SPL meter tomorrow. With two subs, I would assume I would obviously calibrate each one seperately. But after I do that, and I turn both on, and the bass is now louder than the other 5 speakers, would I tweak further?...and actually is that even possible? Better example, if one sub's gain is set to output about 70db, and then I do the same with the other sub, then run both at same time, would output still be 70db? Anything different I would do having two subs? Also, if BD or aramis could link the HSU excel sheet, I'm curious what it is. Each of my 5 speakers have different xover points, so this could be interesting.

Calibrate each sub separately. After that, run both of them together and lower the level until you get the same level as the other speakers.

Use the low pass filter as your crossover to calibrate your subwoofer. The LPF filter should be as high as the highest crossover (high pass filter) you set for your other speakers.

In the bottom of the original post, there are links to two Excel worksheets.

[sbjork]

The three Stereophile Test CDs also have a wide variety of test tones, including various warble tones and sweeps. They are $25 for the set on the Stereophile website and are well worth the purchase.