THD has a much bigger effect on sound than you think

[solderdude]

I am talking about "cumulative" bass distortions. Not a single tone distortion. For example, add all distortions from the bass notes, for example 20 to 200 hz,

The answer is still no the differences between speaker and (worst amp ever measured here) are already too big for this.
IM products is a totally different matter. The harmonics is no issue.

 

[solderdude]

please correct me if I'm wrong!

You are not wrong so no correction is needed .
With huge amounts of IM this might become audible with some specific recordings but not at 0.1% or so.
There are other distortions than just HD and IM though and some are more audible than others.
I believe there is a nice video about this subject somewhere.
crossover distortion for instance is hardly measurable at higher power levels yet can be audible at lower levels.
Measuring amp distortion at lower levels will show this that specific type of distortion though.

 

[Pdxwayne]

The answer is still no the differences between speaker and (worst amp ever measured here) are already too big for this.
IM products is a totally different matter. The harmonics is no issue.

Yeah, hard to separate pure distortions from IM.

My point was the amp could still add a couple of dB to trebles. Depending on the music (overall bass to treble balance), that 2db or so can be audible.

 

[solderdude]

My point was the amp could still add a couple of dB to trebles. Depending on the music (overall bass to treble balance), that 2db or so can be audible.

The answer is no... it won't add even anything near 0.1dB in any part of the frequency band.
To add 2dB in harmonics you need to have 80% THD.
Even when you reach that in the lows caused by the woofer clipping (won't sound nice) you still would not see harmonics in the treble range.

 

[Pdxwayne]

The answer is no... it won't add even anything near 0.1dB in any part of the frequency band.
To add 2dB in harmonics you need to have 80% THD.
Even when you reach that in the lows caused by the woofer clipping (won't sound nice) you still would not see harmonics in the treble range.

Again, I am talking about overall. All the distortions from bass regions should be able to increase the sense of treble by 2db?

For a very rough example, red region combined distortions added to overall yellow region? Can it skew the balance of bass vs treble?

 

[solderdude]

Again, I am talking about overall. All the distortions from bass regions should be able to increase the sense of treble by 2db?

Again... no... harmonics from the bass won't reach those levels in the frequency band you mention.
And certainly not 2dB for that you need higher order harmonics at huge levels.
It does not happen unless you are clipping the amp or speakers.

 

[Pdxwayne]

Again... no... harmonics from the bass won't reach those levels in the frequency band you mention. And certainly not 2dB. It does not happen unless you are clipping the amp or speakers.

Well, yesterday, I was using a web site to calculate the cumulative db value of adding sources of certain dB. Using just 30 sources of -80db, which I assume in bass region, there are much more than 30 fundamental bass sources that would produce distortion of -80db. Adding just 30 sources of -80db quickly add up to additional 2db.

 

[solderdude]

There is only 1 signal per channel. It already consists many added signals with its harmonics.
In the bass part you mentioned there were only 2 notes A0 and C1 that had a relatively high amplitude and most likely not played at the same time but after each other.
These do not have harmonics high enough to reach the treble region.

 

[Hayabusa]

Well, yesterday, I was using a web site to calculate the cumulative db value of adding sources of certain dB. Using just 30 sources of -80db, which I assume in bass region, there are much more than 30 fundamental bass sources that would produce distortion of -80db. Adding just 30 sources of -80db quickly add up to additional 2db.

30 coherent sources of -80 db add up to a signal of -50dB.
this adds 0.026 db to the fundamental..

 

[Pdxwayne]

30 coherent sources of -80 db add up to a signal of -50dB.
this adds 0.026 db to the fundamental..

What if that -50db is added to -65db treble?

 

[solderdude]

30 coherent sources of -80 db add up to a signal of -50dB.
this adds 0.026 db to the fundamental..

Correct... 0.026dB so whatever calculator @Pdxwayne was using is totally wrong.

I'll explain it to Pdxwayne.
reference = 1V
-80dB = 0.1mV
Now here comes the tricky part... we are going to assume all 30 sources of -80dB are in phase (would be absolute worst case scenario).
30x 0.1mV = 3mV
So 1V vs 1.003V in dB is 0.026dB... quite far removed from the fictional 2dB.

What if that -50db is added to -65db treble?

The problem is, and I have been trying to explain this, that for harmonics in the bass to reach treble frequencies they would have to be substantial and very high order. These aren't there so they cannot reach the treble range.
Your 'what if' does not exist in the real world unless caused by clipping.

 

[Hayabusa]

What if that -50db is added to -65db treble?

Then it could be significant

 

[solderdude]

What if that -50db is added to -65db treble?

-50dB of what frequencies (harmonics) and -65dB of what treble frequencies ? Are the fundamentals still there and what frequencies are they ?
When they consist of 30 sources and are also in sync they would also add up.

So.. the answer is a question mark because -50dB of what combined with -65dB of what compared to what reference at what SPL with what masking and which IMD products ?

 

[Pdxwayne]

-50dB of what frequencies (harmonics) and -65dB of what treble frequencies ? Are the fundamentals still there and what frequencies are they ?
When they consist of 30 sources and are also in sync they would also add up.

So.. the answer is a question mark because -50dB of what combined with -65dB of what compared to what reference at what SPL with what masking and which IMD products ?

It becomes complex once you add up masking and IMD....

Anyway, let's use around 1200hz as example for the treble.

Let's assume 1180 to 1220 Hz got lots of harmonic from bass regions, including 5th harmonic (as your can see in your worse amp example, even at 2 watts, 5th harmonic is at -80db). If those bass notes are around 90+db, then those fundamental bass notes won't be loud enough to mask the treble, assuming in ~65db. If indeed those harmonics around 1200 Hz is about -50db, then quite obvious those added much more than 2db to the 1200 area, which is at around -55db.

 

[Blumlein 88]

It becomes complex once you add up masking and IMD....

Anyway, let's use around 1200hz as example for the treble.

Let's assume 1180 to 1220 Hz got lots of harmonic from bass regions, including 5th harmonic (as your can see in your worse amp example, even at 2 watts, 5th harmonic is at -80db). If those bass notes are around 90+db, then those fundamental bass notes won't be loud enough to mask the treble, assuming in ~65db. If indeed those harmonics around 1200 Hz is about -50db, then quite obvious those added much more than 2db to the 1200 area, which is at around -55db.

Are you describing IMD from 20 hz around a tone of 1200 hz? If so, no the fundamental at 20 hz won't mask it, but the 1200 hz tone will. Without the 1200 hz tone to intermodulate around you don't have the distortion from 20 hz notes.

 

[Pdxwayne]

Are you describing IMD from 20 hz around a tone of 1200 hz? If so, no the fundamental at 20 hz won't mask it, but the 1200 hz tone will. Without the 1200 hz tone to intermodulate around you don't have the distortion from 20 hz notes.

No, so far I am talking about just harmonics, like 5th, from around 200hz to 300hz. If you have a high distortion amp, would those harmonics added enough db to around 1200hz area to make it louder than original intented?

 

[pma]

Are you describing IMD from 20 hz around a tone of 1200 hz? If so, no the fundamental at 20 hz won't mask it, but the 1200 hz tone will. Without the 1200 hz tone to intermodulate around you don't have the distortion from 20 hz notes.

You may make any example you want. This is a real world (speaker) measurement, and masking is questionable, especially on diff

frequencies pre-basic frequencies of the excitation tones.

 

[audio2design]

You may make any example you want. This is a real world (speaker) measurement, and masking is questionable, especially on diff frequencies pre-basic frequencies of the excitation tones.

Most modern amplifiers have their lowest distortion at low frequencies (where feedback is maximized). So, where the speaker is worse, amp is best. I don't think you are making an argument that the additive distortion of the amplifiers is meaningful, all this shows is that speakers have a lot of IM. Doppler distortion which reveals as IM is hard to overcome.

 

[Blumlein 88]

You may make any example you want. This is a real world (speaker) measurement, and masking is questionable, especially on diff View attachment 160772frequencies pre-basic frequencies of the excitation tones.

Well in your example the only one likely to be audible is D2. But our hearing is rather insensitive there so it may or may not be. Would depend upon how quiet the location was.

 

[audio2design]

Well in your example the only one likely to be audible is D2. But our hearing is rather insensitive there so it may or may not be. Would depend upon how quiet the location was.

Effectively with loudness contours 60db down from the peak? I am going to go with totally masked.