Speaker Frequency Response - Perception or Reception?

[McGuyver]

So I've noticed that a 10dB SPL difference (going from 90 dB SPL C weighted to 100 dB SPL), at least to my ears in the room, sounds like a huge difference in my speaker cabinet's response.  Sitting at 90 dB SPL the sound becomes very muddy and seems to lose almost all treble.  However, when I turn up the master volume, no other changes to the amp, all the frequencies come shining through again.

Does anyone know if this is a common phenomena with speakers?  Is it just my hearing, and the frequencies are actually still there?  Fortunately I have access to recording equipment, so I will test this theory out and post some results.  I'm thinking it's my speakers' response being different at SPL readings, but I'm willing to bet it's just my ears response.

 

[pabloman]

My guess is there are several factors that may or are contributing to this. It could be the amp/tubes, room acoustics, cabinet construction, speaker construction, cone travel/back pressure. Things could even be situated such that the higher frequencies being more directional than the lower frequencies are just able to be heard more with the increased volume. It is a very real occurrence though.

 

[swarfrat]

No, it's not just YOUR hearing. Its human physiology.
Fletcher Munson curves

 

[AndyG]

So I've noticed that a 10dB SPL difference (going from 90 dB SPL C weighted to 100 dB SPL), at least to my ears in the room, sounds like a huge difference in my speaker cabinet's response.  Sitting at 90 dB SPL the sound becomes very muddy and seems to lose almost all treble.  However, when I turn up the master volume, no other changes to the amp, all the frequencies come shining through again.

Does anyone know if this is a common phenomena with speakers?  Is it just my hearing, and the frequencies are actually still there?  Fortunately I have access to recording equipment, so I will test this theory out and post some results.  I'm thinking it's my speakers' response being different at SPL readings, but I'm willing to bet it's just my ears response.

Human hearing is a strange thing ... for most people, (and according to Fletcher and Munson), human hearing is most "flat" at 85db ... we perceive most frequencies equally.  Turn it up louder, and we become more sensitive to high and low frequencies ... almost like putting a "happy face" curve on a graphic EQ.  At lower volumes, we are most sensitive to midrange (where most frequencies in nature occur ... chalk that one up to Darwin).
There is also no such thing as a flat full-range speaker.  Studio monitors can be tuned to a certain degree, because they have at least 2 drivers (bass and treble), and the cabinet can be matched acoustically to those drivers.  Guitar and bass cabinet speakers, generally, are a compromise to get the best sound out of a single cone.
When you factor in the room you're playing in, things get even more complicated.  High frequencies are more directional, while bass frequencies go everywhere (again, generally speaking).  My guess is your perception is more of an acoustic phenomenon than something the speakers are actually doing.

 

[Cagey]

So I've noticed that a 10dB SPL difference (going from 90 dB SPL C weighted to 100 dB SPL), at least to my ears in the room, sounds like a huge difference in my speaker cabinet's response.  Sitting at 90 dB SPL the sound becomes very muddy and seems to lose almost all treble.  However, when I turn up the master volume, no other changes to the amp, all the frequencies come shining through again.

Does anyone know if this is a common phenomena with speakers?  Is it just my hearing, and the frequencies are actually still there?  Fortunately I have access to recording equipment, so I will test this theory out and post some results.  I'm thinking it's my speakers' response being different at SPL readings, but I'm willing to bet it's just my ears response.

It's part physiology, as has been presented, and part physics. If a speaker cone is being told to move back and forth at 500hz, then another control signal comes along at the same time telling it to move back and forth at 5000hz, what's a mother to do? It can only move at one frequency or the other. Some superimposition can take place, but only so much. One or the other has to take a back seat. That's why there are multiple speakers of different size in full-range units. Using filters (crossovers) you can send the appropriate task to the appropriate device.

It's been studied for years in the high fidelity sound reproduction realm to great effect, but only in the last decade or so as it applies to original sound production. Especially now with some of the high-end modelers out there. Demand is going up for full-range flat-response speakers with wide dispersion angles. As a result, we're seeing smaller and more accurate PA systems, monitors and guitar speakers.

 

[drewfx]

If a speaker cone is being told to move back and forth at 500hz, then another control signal comes along at the same time telling it to move back and forth at 5000hz, what's a mother to do? It can only move at one frequency or the other. Some superimposition can take place, but only so much. One or the other has to take a back seat.

Um, no.

The frequencies add and you just get a more complex waveform, simple as that. It's more difficult to cleanly produce a single frequency at the extreme than a very complex waveform with all of its frequency content below the extreme.

 

[Volitions Advocate]

Actually it's impossible to produce only one frequency. The overtone series is present in pretty much every tone you hear.. The closest thing in nature you can get to a pure sine wave would be a tuning fork.

In essence what you've both said is correct, but I could also be misinterpreting your comments based on my own bias.

Also. in regards to sound levels.  It's important to remember just how much of a difference 10 dB makes.  Every 6 dB is a doubling of sound amplitude.  So its a logarithmic scale.

Plus it's all subjective with a mediated point of origin based on consensus.  One point where a lot of people make a mistake is in believing the Fletcher Munson curves are a scientific fact, they aren't.  It's just an experiment based on subjective responses from people they were blasting with sound.  The curves are an average of those responses.  It may amount to a very small margin of error, but still its the truth.

Everything about this science is based on some reference value.  The fletcher munson curves, even the marking on your speakers.  SPL is sound pressure level, but its based on a reference value and goes up or down from there.  The reference could be subjective in the case of Fletcher and Munson, or variable depending on the air pressure in your neck of the woods. Or barometric pressure or whatever.  There are half a dozen ways to measure how many dB's you're pushing.

I realize I'm deviating from the original question, probably.  But we could go deep into these things and come up with some crazy reason.  Fourier theory is a good place to pick away at it too, especially when you're talking about frequency response and the overtone series making complex waveforms.

 

[AndyG]

Plus it's all subjective with a mediated point of origin based on consensus.  One point where a lot of people make a mistake is in believing the Fletcher Munson curves are a scientific fact, they aren't.  It's just an experiment based on subjective responses from people they were blasting with sound.  The curves are an average of those responses.  It may amount to a very small margin of error, but still its the truth.

The Fletcher-Munson curves, or equal-loundness curves, which were both derived from a series of experiments, show a consistant trend within the human population that were tested, which makes the information factual.  There will be deviations in response from person to person, but as a general rule, they show how human hearing responds to different sound pressure levels.
Just like everything else that is human, variations obviously occur.  Same thing for anything "percieved".  The body's reacton to temperature, for example ... we here in Canada will walk around wearing shorts at the same temperature that people in the Southern US would probably be more comfortable wearing a sweater ... even though the temperature remains the same.  If you show 100 people a purple card, some will see it as being violet, and others mauve.  People can "learn" to discern frequency response at any volume ... that's why really good mixers and producers create albums that sound great at any volume ... but the way the human ear works remains unchanged, and that is what Fletcher and Munson were trying to determine.

 

[drewfx]

Actually it's impossible to produce only one frequency. The overtone series is present in pretty much every tone you hear.. The closest thing in nature you can get to a pure sine wave would be a tuning fork.

If you're playing back a signal through a loudspeaker, it doesn't matter what "the closest thing in nature" is. You can certainly generate (something very close to) a pure sine wave and send it to the speaker. You can also take a "natural" complex waveform and filter it to get it closer to a pure sine wave.

Plus it's all subjective with a mediated point of origin based on consensus.  One point where a lot of people make a mistake is in believing the Fletcher Munson curves are a scientific fact, they aren't.  It's just an experiment based on subjective responses from people they were blasting with sound.  The curves are an average of those responses.  It may amount to a very small margin of error, but still its the truth.

The scientific fact of equal loudness contours (though Fletcher-Munson is commonly cited, ISO 226:2003 is the much more recent and accurate version) is that (to my recollection) no one tested does better than a few dB of the curves. This is not surprising when you consider that the varying sensitivity at different levels and frequencies occurs because of physics and the underlying physiology of human hearing.

And gravity is not 100% constant everywhere on Earth either. Does that mean it isn't a "scientific fact"?

Everything about this science is based on some reference value.  The fletcher munson curves, even the marking on your speakers.  SPL is sound pressure level, but its based on a reference value and goes up or down from there.  The reference could be subjective in the case of Fletcher and Munson, or variable depending on the air pressure in your neck of the woods. Or barometric pressure or whatever.  There are half a dozen ways to measure how many dB's you're pushing.

Be careful. A decibel is a logarithmic expression of a ratio.

If "dB" with no suffix is used, it means 2 values are being compared, and the number of dB's is expressing the ratio between them.

If "dB with a suffix" is used, the suffix establishes a reference level and only one value is being compared to that reference level.

So "dB SPL" by definition is always comparing a measured Sound Pressure Level to the reference level of 0dB SPL = 0.00002 Pa.

 

[dudesweet157]

I like shakey cheez on my bisketti. 

Also, speakers sound mo' bettah at high volume (until they start to clip) because the Flying Bisketti Monstar wills it so!!!