Calibration dispute !

[Vaughan Odendaa]

Hi there,

I had a bit of a dispute between a collegue of mine at work today. My manager has 30 years of experience, he is an audio engineer, a ton of certificates, etc.

We were basically talking about calibrating speakers and using the SPL meter to calibrate. I just hooked up an amp today at work and he was using his ear to calibrate. And so I told him that to be sure, I'll bring in my sound pressure meter (Ratshack) to confirm that the levels are even.

He said to me that the tones used in the reciever may be more sophisticated than the SPL I'm using to measure with. I don't know what you guys think of that.

Strange.

He also told me that I can't measure frequencies with the SPL meter. Only gain. Which I know. But the wording comes out wrong. The meter can't actually measure frequencies, but it measures the output at those frequencies the speakers or subwoofers are playing.

And so I said that if the speakers are playing back, let's say, 50hz, I can measure the SPL at that frequency. From there, he tells me, "how do you know that you are getting 50 hz "out", compared to what the amplifier puts in. My boss then comes in and tells me, how do you know that you are getting 50 hz out, as well, because there is loss involved in the amplifier chain.

I simply looked blank because I don't quite understand what he mean't by that. I have never been asked that question and I, to be honest, have never encountered that question on any audio forum. I thought he was nuts ! He tells me that the SPL meter only measures noise and the resulting reflections, but not the actual amplifier input.

Again, I have no clue. My boss then comes into the discussion and tells me that the pink noise that is used for subwoofer calibration is relatively high frequencies mixed with some low frequencies (around 50 hz) which I think is correct. But then he made the abnormal statement that the SPL can't calibrate the subwoofer down very low. He tells me, "what about very low frequencies in the built-in test tones ? He says "it doesn't reach down to below 40 hz

Huh ?

Pink noise is what is used to calibrate subwoofers, correct ? Is pink noise the standard set for subwoofers by the SMPTE ? I just need some confirmation on this because I was asked questions which, to me, just seem ridiculous.

You know, I went to the very basics and told me manager, "if you use a test tone disk with tones ranging from 40 hz to 200 hz, you can measure individual tones. My manager tells me, "the SPL CANNOT measure frequency". Only a spectral analyser can !
"
But an SPL can measure the gain at that frequency. "No it cannot", he proclaims.

Whew. So at towards the end of this dispute, my manager used his 30 years of experience drawcard to invalidate what I said to him. This is just so frustrating. When you know something as basic as this and it gets skewed up so badly.

What I didn't understand is what he was talking about in terms of measuring the amplifier input level (he tells me that we can only measure the output level), and he says how do you know that the input measures to the output.

So I just wanted to know from you guys what you think about this. Perhaps you can give me a better understanding of this whole issue.

 

[Audioholics]

Most pink noise isn't accurate on the subwoofer test tones output by receivers. In addition, the Radio Shack meter isn't very accurate at lower frequencies...

Some of the other stuff I read, assuming you worded it correctly, sounded like mumbo jumbo from your boss, however.

Bottom line is that a Rat Shack meter and receiver test tones (pink noise) is much better than your ear for calibrating the main 5-7 speakers in a home theater or two-channel system. Unless you are the bionic-woman. For the subs, you may have to use your ear or a more sophisticated tool and measure various frequencies... the key here being to better integrate the sub and deal with room modes and dips.

 

[BigGuy]

From what I have learned from experience and a lot of reading, The Ratshack SPL is a great tool for 50$, but it's accuracy is only down to +-2 db, which is great for a 50$ meter, it has very good repeatability, whish is more important for level calibration (who cares what the number is, as long as all the numbers are the same). C weighting will take into account 32hz to 10000 hz, and A is 500 to 10000 hz, C would be better for the sub woofer calibration and A for all the other speakers (I think I have those numbers right). I use mine, and get very similar numbers to the auto calibration when I had a Denon 3805 (so it has to be somewhat close!) As far as I know, if you play a specific tone, lets say 1khz and measure the SPL with the RatShack meter, and then compare the SPL to another frequence, like 500hz,and so on, you could plot out a frequency response. So I would think that your logic would be correct as far as the use of the meter. My boss is the same way, he doesn't even adjust any settings on the projectors or pre\pros. He says that you have to let it break in for a few months before it's even worth fooling with... I don't think he's right about that but one thing that I've learned is that the boss is always right, no matter how wrong he is... no point of trying to convince him otherwise! lol.

Brian

 

[Vaughan Odendaa]

The manager is the guy that is giving me all the grief. Perhaps someone could address the other points I made concerning measuring the input to the output.

But what my boss said about calibrating subwoofers down to 30 hz is pure and total nonsense. Please correct me if I'm wrong but the SMPTE standard uses pink noise to test subwoofers world wide. It's a standard that is used.

Pink noise doesn't generally have 30 hz information, correct ? It's more like a combination of various frequencies. It's the mixing engineers job to "calibrate" the recording so that 40 hz will be used at the right moments and at the right levels.

Because he tells me that if I calibrate the subwoofer to 75 dB's, that's using pink noise. What about frequencies not used in the pink noise ? What about them, he asks me.

Perhaps someone more qualified than I can address these finer points. Thanks.

--Sincerely,

 

[billnchristy]

from an online dictionary:

" White noise is a sound that contains every frequency within the range of human hearing (generally from 20 hertz to 20 kHz) in equal amounts. Most people perceive this sound as having more high-frequency content than low, but this is not the case. This perception occurs because each successive octave has twice as many frequencies as the one preceding it. For example, from 100 Hz to 200 Hz, there are one hundred discrete frequencies. In the next octave (from 200 Hz to 400 Hz), there are two hundred frequencies.

White noise can be generated on a sound synthesizer. Sound designers can use this sound, with some processing and filtering, to create a multitude of effects such as wind, surf, space whooshes, and rumbles.

Pink noise is a variant of white noise. Pink noise is white noise that has been filtered to reduce the volume at each octave. This is done to compensate for the increase in the number of frequencies per octave. Each octave is reduced by 6 decibels, resulting in a noise sound wave that has equal energy at every octave. "

 

[mtrycrafts]

The manager is the guy that is giving me all the grief. Perhaps someone could address the other points I made concerning measuring the input to the output.

But what my boss said about calibrating subwoofers down to 30 hz is pure and total nonsense. Please correct me if I'm wrong but the SMPTE standard uses pink noise to test subwoofers world wide. It's a standard that is used.

Pink noise doesn't generally have 30 hz information, correct ? It's more like a combination of various frequencies. It's the mixing engineers job to "calibrate" the recording so that 40 hz will be used at the right moments and at the right levels.

Because he tells me that if I calibrate the subwoofer to 75 dB's, that's using pink noise. What about frequencies not used in the pink noise ? What about them, he asks me.

Perhaps someone more qualified than I can address these finer points. Thanks.

--Sincerely,

Perhaps WmAx will stop in to comment?
But, your boss, from reading your account seems to be wrong on a number of issues, 30 years of bad experience No, 30 years of experience is not a guarantee that he knows what he is talking about. If that was the case, then the world would be a much better place for everything. It is far from it
While it doesn't address your question, I had a similar encounter with a master plumber who even taught classes Clueless!!!

What are you guys calibrating? Level matching or EQing the speaker systems?

If he want the high frequencies calibrated, it is not done through the sub and adjusting the gain of the low band on an EQ would not affect the high band. Besides, the crossover to the sub should minimize the frequencies above that point, and is not part of the pink noise.
As is pointed out in the post below by the definition explanation, depending on the pink noise, as some are 1/3 octave types, it would contain all the frequencies, but a sub crossover would limit the frequencies above the filter by the slope of the crossover. NO, the 30Hz frequency would not contain any 200Hz signals, or 1kHz signal. He is mistaken.
The spl meter reads in sound pressure levels whether it is a single tone pure frequency, 1 kHz, or combines it all and averages it.
SPL meter is not the tool to use to EQ anything as it doesn't separate out any frequency. How could it with one meter readout? But, a RTA could for the designed frequency or band, or software specifically made to EQ speaker setups with swept sine waves from 20Hz to 20kHz.
Going back to the bosses belief about high frequencies, that would be EQ not through the sub but the other speakers that handle those bands

His ears will not detect single frequency issues in a pink noise signal. And, his ability to level them by ear would be no better than 1dB spl at best in mid band and much worse in the two ends.

Your boss should enroll in some acoustic classes, not OJT over 30 years as he picked up some nonsense, it appears. But, you may be looking for another job if you disagree???

 

[MDS]

I'll answer this from a slightly different perspective than the other posters. This is something the boss should know cold but apparently does not.

Any sound you hear, especially music, is made up of MANY waves of different frequencies and amplitudes all superimposed on top of one another. It's why you can hear the doorbell when the music is playing! You ear does not filter out any specific frequency nor does the SPL meter. An SPL meter is not the right tool for the job of measuring specific frequencies.

Use an audio editor and create two different waveforms, say one 40 Hz at 0 dB and another say 10 kHz at 0 dB and mix them together. Result - ONE waveform that contains both. Your ears hear the combined wave as one sound. The pink noise (see definition above) is the same concept except there are many more than two component waveforms.

 

[agarwalro]

"how do you know that you are getting 50 hz "out"

I believe that is a valid concern. Here is one way,

1) Synthetically generate tones at desired frequency.
2) Use transport with FR tested to be 5 - 25,000Hz +\- 0.00000Hz
3) Use microphone with FR tested to be 5 - 25,000Hz +\- 0.00000Hz
4) Use spectrun analyzer.

And as they say... WYSWYG

 

[vierling]

Human Hearing & Pink Noise

First of all: Human ear does not have a flat freq response.
Our freq response changes with more or less sound pressure.

For very soft sounds, near the threshold of hearing, the ear strongly discriminates agains low frequencies. For mid-range sounds around 60 phons, the discrimination is not so pronounced and for very loud sounds in the neighborhood of 120 phons, the hearing response is more nearly flat.

One of the implications of this aspect of human hearing is that you will perceive a progressive loss of bass frequencies as a given sound becomes softer and softer. For example if you are listening to a recording of an orchestra and you turn the volume down, you will find that the bass instruments are less and less prominent. This is the purpose of the so-called "loudness contours" on audio amplifiers; they allow you to boost the bass frequencies when you are listening at low sound levels to give you a more realistic balance of the high and low frequencies in the music.

If You are using an spl meter to calibrate, You want to use pink noise, because music has to many freq´s peaking at different levels. music, ( Kick drum, snare, guitar... )

IN THE SPL METER:
SETTING THE WEIGHTINGWEIGHTING lets you set the meter’sweighting characteristics, which deter-mine the meter’s frequency responsecurve. You can select A-weighting or C-weighting.A-weighting has A-curve frequency characteristics. This setting causes the meter to respond mainly to frequencies in the 500-to-10,000 Hz range, which is the human ear’s most sensitive range.
Select A-weighting to determine an area’s noise-level. Take measurements at several dif-ferent points in the area.

C-weighting has C-curve (flat) frequency characteristics. This setting causes the meter to respond mainly to frequencies in the 32 to 10,000 Hz range. Select C-weighting to measure sound levels of mu-sical material or calibrate sound systems.

PINK NOISE
For processes of testing and equalizing rooms and auditoriums, it is convenient to have broad-band noise signals. Typically, white noise or pink noise is used. Whereas white noise is defined as sound with equal power per Hz in frequency, pink noise is filtered to give equal power per octave or equal power per 1/3 octave. Since the number of Hz in each successive octave increases by two, this means the power of pink noise per Hz of bandwidth decreases by a factor of two or 3 dB per octave.

Since pink noise has relatively more bass than white noise, it sounds more like the roar of a waterfall than like the higher hissing sound of white noise, and seems to be the best option when You want to correct Human ear freq response with a sound system.

Pink noise is often the choice for equalizing auditoriums. Real-time analyzers can be set up so that they display a straight horizontal line when they receive pink noise.

 

[mtrycrafts]

An SPL meter is not the right tool for the job of measuring specific frequencies.

Certainly cannot separate out frequencies to measure in a group of frequencies. But, it can measure single pure tones by themselves, say 1kHz.

 

[Vaughan Odendaa]

But my manager tells me, no, no, NO, "a SPL meter cannot measure frequency !". "It can only measure gain at that frequency", he proclaims.

We went back and forth on this when we had the dispute. But what I didn't understand is when he said, "how do you know that the input is accurate with the output and that the SPL meter can read the input as accurately as the output.

I don't know. Perhaps someone will chime on that.

As to my boss's suggestion that the sub tone (I think it's a warble tone with lots of low bass tones merged into one) does not contain very low bass and that you wouldn't be able to calibrate at frequencies lower than what is used in recievers for subwoofers.

Perhaps someone could just add a comment on that too. Doesn't the SMPTE state that pink noise is used to calibrate with speakers worldwide and with subwoofers ?

This whole issue is an SMPTE thing. Although he did mess up at one point in the argument because he said that only run-of-the-mill recievers have test tone generators. Ahem, except the TagMclaren AV192 also has test tone generators.

Ouch.

My manager is the more knowledgable of the two. But he doesn't seem to understand simple concepts. Telling me that an SPL can't measure frequency even though I said "it can measure the SPL at different frequencies".

So his case is that the SPL can measure gain but it cannot measure frequency. I still don't understand what he mean't with measuring the input to the output.

Because he said on a few occasions that the SPL meter only reads the outside noise, the reflections, everything, but what about the input in the amplifier. How does the SPL know what the input is from the output and how do you know that it's accurate.

This is what he said.

Perhaps someone could explain this to me. But at the end of the day, he said that the SPL is not accurate. I have used the Radio Shack meter for almost a year and I consider it to be one of the most valuable tools I own.

--Sincerely,

 

[Vaughan Odendaa]

And one more thing. The sine waves used on test tone cd's. They only contain a fundamental wave, correct ? Like a 30 hz tone. There are no harmonics or overtones involved there.

Correct ?

--Sincerely,

 

[BMXTRIX]

Your boss is correct. If you feed an amplifier/receiver a 1kHz tone, you will need very specialized equipment to actually measure and find out that you are in fact receiving a 1kHz tone from the speakers. You can get whatever tone comes out of the speakers and put it in line with the other tons that come from the speakers, but there is no way to know for sure what the tone is.

Just because you feed it what you believe is a 1kHz tone (are you sure?) and just because you think you are hearing a 1kHz tone... you can't be sure other than to be hopeful.

I am not sure how much colorization of specific tones that most receivers have, but I believe that when you hear reviews of receivers being 'warm' or 'neutral' they are talking a bit about the tendency of receivers to shift frequencies a bit.

EITHER WAY: You should use a SPL meter to setup the SPL of a receiver. Actual equalization is completely different and is often done by ear since the right gear is not cheap - or common. It also is very much personal preference.

 

[MDS]

But my manager tells me, no, no, NO, "a SPL meter cannot measure frequency !". "It can only measure gain at that frequency", he proclaims.

I think I see where he is coming from. The SPL meter measures sound pressure level so it is true that it cannot 'measure frequency'; in other words you cannot play a tone and have the meter read '120 Hz'. Measuring frequency is not what it is designed to do in the same way that a crescent wrench wasn't designed to put screws in the wall.

The meter is 'listening' to ALL sounds in the room which is why it is important to do the calibration in a quiet environment, avoid reflections from your body, etc. But the purpose is simply to make the sound pressure level roughly equal from all speakers. It matters not that the meter isn't accurate to within .1 dB - it matters that it is repeatable.

We went back and forth on this when we had the dispute. But what I didn't understand is when he said, "how do you know that the input is accurate with the output and that the SPL meter can read the input as accurately as the output.

Here is my take on what he is getting at:
He is trying to correlate the input signal with the output; ie if you play a pure tone of 120 Hz from a test CD and the amplifier amplifies it, will the tone produced by the speakers and heard by your ears still be the pure 120 Hz tone? It will if the amplifier is PERFECT and the speakers are PERFECT which of course will not happen in the real world. That has nothing to do with using an SPL meter to measure the sound pressure level. The meter is not reading the input and the output consists of all the sound produced by the speaker and all other sounds present in the room. Your purpose is simply to adjust the gain of each channel so that when you play the tone at a given volume setting the output SPL reaches your target.

Pink noise is used because it is wide bandwidth general purpose 'noise'. You wouldn't want to calibrate with a specific musical selection because your calibration will only be accurate for that selection.

Here's the key point I try to drive home every time there is a discussion on calibration and 'reference levels':
When you calibrate with pink noise, the average (RMS) level of the signal is at some value (-30 dB for internal test tones). When you adjust the channel trims so that the meter reads 75 dB, you have Dolby Reference Level, BUT the meter will only read 75 dB when the source you play has an average level of -30 dB. Play a CD with an average level of -10 dB and the meter will read MUCH higher (obviously an average of 20 dB higher.). -30 dB is a THX standard and was chosen because it just so happens that -31 dB is the 'standard' for average level of dialog in a movie. Using that test tone and doing the calibration gets you a pretty good all around ballpark figure for the majority of movies if you like pain and actually want to listen to the movie at reference level. Music will be much louder at the same volume setting because the average level of music on a CD is much higher than the level you used for doing the calibration.

 

[Vaughan Odendaa]

I think I understand. I thought that it was obvious that a 30 hz recorded tone played through the speakers would still be a 30 hz tone. But if it isn't, then why bother recording these frequencies on test disks ?

If no one knows for sure if the frequency will be "retained" from input to output, then what does that mean for us ? That nothing is actually accurate at all ?

So if I play a 50 hz test tone, if may be something else ? This is what I was confused about. I have never heard of anyone on any forum make this kind of statement before because it just seems so very unusual, strange even.

Perhaps you can straighten this out for me. Oh, yes, another thing I remember my manager telling me is "how does the SPL differentiate between the fundamental and the harmonics ?". "How does it measure harmonics ?"

A 30 hz tone does not contain any harmonics, correct ? It is just a simple wave. A single tone with no overtones. Correct ?

--Sincerely,

 

[jaxvon]

Well yes, the tone on the CD is 30Hz. Even after passing through your amp it's basically a 30Hz tone with few harmonics. However, after it comes out of your speakers, there are going to be some other harmonics added. Most speakers have single to double digit distortion figures at 30Hz. For example, a Velodyne DD-12 has about 3%THD playing 30Hz at 90dB [Source: AV Talk Subwoofer Tests].

 

[Vaughan Odendaa]

Yes, but this depends on output. But at what point does this become audible ? I mean, 2% THD doesn't, to me, seem like something to worry about. 10% THD, on the other hand, is something else.

So with a 30 hz tone passing through the amp to the speakers, it will largely still be a 30 hz tone. At reasonable levels, there should be no overtones present but at extreme output levels, harmonics will be added from the speakers (and amp).

But is this a point of concern ? That we don't know, as a poster suggested, that a tone from the amp will not be exactly the same on the output end ?

If I put in a 30 hz I expect to measure output at 30 hz.

--Sincerely,

 

[jaxvon]

I still think it's a moot point. Even if it's not a "pure 30hz", it's what you're going to get when *any* 30Hz tone is playing. It's what you have to work with. You can't try to calibrate for an imaginary system, you can only calibrate what you have.

But yes, at such low frequencies, the THD needs to get well into the double digits to be audible.

 

[Vaughan Odendaa]

So how does one get over the argument that the SPL cannot measure the input accurately with the output ? It seems like it could be a valid argument and I never thought about it until he mentioned it.

Then again, he is assuming that it won't track the input to the output properly. Just remember that this guy is an EE. But he asked me questions that I don't think people encounter on these forums everyday.

As a matter of interest, how accurately does the SPL meter track the input to the output ? Does anyone know ? Since he brought up these questions in the first place, I am a little interested to know.

Thanks.

--Sincerely,

 

[mtrycrafts]

I am not sure how much colorization of specific tones that most receivers have, but I believe that when you hear reviews of receivers being 'warm' or 'neutral' they are talking a bit about the tendency of receivers to shift frequencies a bit.
.

I believe that those terms are used to indicate a reduction of some frequency band volume to get a 'warm' sound, not a shift. How would one know if the frequency shifted without some precise reference to compare to.

Also, I just don't see how a speaker would not reproduce a voltage changed at 1kHz and output one that is 1.01kHz. Or worse, further off. The two magnetic fields react quite well. I am not sure this is what the poster and boss are discussing though in a recording studio but EQ and level matching perhaps.