Can you hear a 1kHz sine wave distortion of 0.85% from tube preamplifier

[RayDunzl]

For fooling around with percents or decibels of harmonics at home::

REW can add harmonic distortion up to the 9th harmonic in any combination and phase angle you desire from the Sine Wave Tone Generator.

Turn the harmonics selected on/off with the "add harmonic distortion" checkbox.

Click a few times (eyes closed) to lose track if it is on or off and try to decide which it is for small defects.

---

It is possible to do little parlor tricks like cancel existing harmonic distortion by adding a harmonic with a phase angle that cancels the existing "real" harmonic (for example, as in a speaker output of what should be a pure tone)...

See https://www.audiosciencereview.com/...than-any-distortion.15880/page-22#post-529367

 

[wwenze]

If it is due to second harmonic being audible, could we adjust our room such that 2nd harmonic is peaking while fundamental is cancelling?

Sidenote: In doing so I found a position where the sound almost fully cancels out but the initial wavefront doesn't get cancelled so it goes loud then soft again each time I press play. Weird.
Sidenote2: I actually found a position that totally cancels. Wow. Pity I lost that position almost immediately before finding if it is a peak for 2kHz or not.

 

[pma]

Thanks to all who are posting the comments and especially the ABX reports! This test may be good for "tuning" oneself to become familiar with caveats of ABX testing. After downloading the files there is a plenty of time for training.

 

[Jimbob54]

Thanks to all who are posting the comments and especially the ABX reports! This test may be good for "tuning" oneself to become familiar with caveats of ABX testing. After downloading the files there is a plenty of time for training.

Daft question about Foobar ABX . A and B stay the same throughout the 16 (or however many) tests you do, dont they? So once you are familiar with them , you just need to listen to X and Y for each subsequent trial? I'm pretty sure that is how it goes and certainly the last few of my trials I was only listening to X and Y (or maybe just X even)

 

[artburda]

Daft question about Foobar ABX . A and B stay the same throughout the 16 (or however many) tests you do, dont they? So once you are familiar with them , you just need to listen to X and Y for each subsequent trial? I'm pretty sure that is how it goes and certainly the last few of my trials I was only listening to X and Y (or maybe just X even)

Exactly

 

[wwenze]

Ok so in the end I found a location where the difference is very obvious, it is as fast as playing two files that are different, didn't even need to listen to the other tracks in each trial. The fundamental does not need to completely cancel; the 2nd harmonic will peak at a position where the fundamental does not fully cancel anyway.

I guess in short, this means finding the location of the 2kHz peak is the key.

Spoiler

foo_abx 2.0.6d report
foobar2000 v1.6.7
2021-09-04 22:19:21

File A: sample1.flac
SHA1: c7fa3903f280279e07fd8a7981017b0634969711
File B: sample2.flac
SHA1: 6f45f66aef302ed42b9199074fbfaca1ab4dd708

Output:
Default : Primary Sound Driver
Crossfading: NO

22:19:21 : Test started.
22:19:25 : 01/01
22:19:30 : 02/02
22:19:36 : 03/03
22:19:40 : 04/04
22:19:44 : 05/05
22:19:48 : 06/06
22:19:50 : 07/07
22:19:54 : 08/08
22:19:59 : 09/09
22:20:02 : 10/10
22:20:08 : 11/11
22:20:12 : 12/12
22:20:16 : 13/13
22:20:19 : 14/14
22:20:22 : 15/15
22:20:27 : 16/16
22:20:27 : Test finished.

----------
Total: 16/16
p-value: 0 (0%)

-- signature --
c60192e92739b46dbf0da40b3c3a7226c26d4c2b

I suggest those who can't hear the difference try searching for this location. Back at normal listening position, I still can't hear the harmonic at all.

I'm still going to leave my response as "cannot hear" due to the conditionality.

 

[levimax]

Ok so in the end I found a location where the difference is very obvious, it is as fast as playing two files that are different, didn't even need to listen to the other tracks in each trial. The fundamental does not need to completely cancel; the 2nd harmonic will peak at a position where the fundamental does not fully cancel anyway.

I guess in short, this means finding the location of the 2kHz peak is the key.

Spoiler

foo_abx 2.0.6d report
foobar2000 v1.6.7
2021-09-04 22:19:21

File A: sample1.flac
SHA1: c7fa3903f280279e07fd8a7981017b0634969711
File B: sample2.flac
SHA1: 6f45f66aef302ed42b9199074fbfaca1ab4dd708

Output:
Default : Primary Sound Driver
Crossfading: NO

22:19:21 : Test started.
22:19:25 : 01/01
22:19:30 : 02/02
22:19:36 : 03/03
22:19:40 : 04/04
22:19:44 : 05/05
22:19:48 : 06/06
22:19:50 : 07/07
22:19:54 : 08/08
22:19:59 : 09/09
22:20:02 : 10/10
22:20:08 : 11/11
22:20:12 : 12/12
22:20:16 : 13/13
22:20:19 : 14/14
22:20:22 : 15/15
22:20:27 : 16/16
22:20:27 : Test finished.

----------
Total: 16/16
p-value: 0 (0%)

-- signature --
c60192e92739b46dbf0da40b3c3a7226c26d4c2b

I suggest those who can't hear the difference try searching for this location. Back at normal listening position, I still can't hear the harmonic at all.

I'm still going to leave my response as "cannot hear" due to the conditionality.

Sometimes tests bring up more questions than they answer. At first it was hard but I did manage to ABX the two files. Then I read this "location" post and realized all I had to do was move my head a few inches one way or the other and I could clearly hear the distortion in the distorted file. Once I learned that I could ABX without even listening to both files.

foo_abx 2.0.6d report
foobar2000 v1.6.7
2021-09-04 08:42:06

File A: sample1.flac
SHA1: c7fa3903f280279e07fd8a7981017b0634969711
Gain adjustment: -13.91 dB
File B: sample2.flac
SHA1: 6f45f66aef302ed42b9199074fbfaca1ab4dd708
Gain adjustment: -13.87 dB

Output:
Default : Speakers (XMOS XS1-U8 MFA (ST))
Crossfading: NO

08:42:06 : Test started.
08:42:31 : 01/01
08:42:46 : 02/02
08:43:12 : 03/03
08:43:22 : 04/04
08:43:56 : 05/05
08:44:31 : 06/06
08:46:34 : 07/07
08:47:18 : 08/08
08:47:18 : Test finished.

----------
Total: 8/8
p-value: 0.0039 (0.39%)

-- signature --
13259539e33712a02e91f1b96d663d4d9898318c

 

[RayDunzl]

If it is due to second harmonic being audible, could we adjust our room such that 2nd harmonic is peaking while fundamental is cancelling?

I'd vote "unlikely" for any condition other than a demonstration that it is possible...

Do a slow sine sweep, monitor with RTA, note how the harmonic levels vary among each other as the fundamental frequency changes.

Now what?

 

[wwenze]

It explains why if given everyone has the same ability to hear, around less than half of people will not hear it based on pure luck of their speaker positioning alone when the harmonic's level drops below a certain point, which is exactly what we are seeing in the poll results, that's what.

Perhaps a better follow-up question would be, can you hear it with headphones. That removes room luck out of the equation, altho I'm not sure about other sources of luck, perhaps headphone measurers can comment?

 

[MakeMineVinyl]

I didn't have time to do a formal test but I listened to both tones from my computer with the speakers about 12 feet to my left. I could hear the one with the distortion reliably, but the difference was very subtle. If I didn't know I was listening specifically for distortion and didn't have two files to compare, I doubt I would notice it. I checked the file in Adobe Audition, and confirmed I was right.

 

[levimax]

It explains why if given everyone has the same ability to hear, around less than half of people will not hear it based on pure luck of their speaker positioning alone when the harmonic's level drops below a certain point, which is exactly what we are seeing in the poll results, that's what.

Perhaps a better follow-up question would be, can you hear it with headphones. That removes room luck out of the equation, altho I'm not sure about other sources of luck, perhaps headphone measurers can comment?

I just tried it with headphones and it was an entirely different experience. After a little practice I could still ABX the two tones but it was definitely harder .... all I could hear was a little "roughness" or "edge" on the distorted track and I had to turn it down and concentrate to hear it. With speakers if I moved to the right spot I could clearly hear the second tone ..... probably some cancelling of the primary tone allowed the distortion to come through loud and clear. Very interesting experience I never would have thought of.

 

[xaviescacs]

It explains why if given everyone has the same ability to hear, around less than half of people will not hear it based on pure luck of their speaker positioning alone when the harmonic's level drops below a certain point, which is exactly what we are seeing in the poll results, that's what.

Perhaps a better follow-up question would be, can you hear it with headphones. That removes room luck out of the equation, altho I'm not sure about other sources of luck, perhaps headphone measurers can comment?

I've just edited my results to add the headphones I used for the test.

 

[pma]

I just tried it with headphones and it was an entirely different experience. After a little practice I could still ABX the two tones but it was definitely harder .... all I could hear was a little "roughness" or "edge" on the distorted track and I had to turn it down and concentrate to hear it. With speakers if I moved to the right spot I could clearly hear the second tone ..... probably some cancelling of the primary tone allowed the distortion to come through loud and clear. Very interesting experience I never would have thought of.

This is the plot that shows a difference between the pure tone (sample2) and tube distorted tone (sample1). The difference file was generated with DeltaWave software by @pkane . We can see that the difference is a 2kHz tone, second harmonic of fundamental 1kHz. If the level is reasonably lowered, we can hear it as a second tone in sample1 even with headphones, in case they do not have too high intrinsic distortion.

I am also posting the sound difference file, shorted to 10s duration, as a zipped flac below. One of the reasons of the thread is to show that hearing is important, not only seeing the plots. To make an interpretation of plots and graphs shown in the ASR reviews is uneasy and deep experience is needed to "translate" them properly.

 

[Pdxwayne]

Topping E30 to L30 to AKG K371

 

[Pdxwayne]

BTW, either my right ear is more sensitive, my chain has a little off balance issue, or the recording again has slight imbalance. I heard the distortion more clearly with my right ear.

 

[antcollinet]

I know why I can't hear it. Tinnitus. It's masking it. When I listended to a 500 Hz tone with -40dB 2nd harmonic I could hear it, but with 1kHz fundamental and higher up I start to hear the distortion when it's -30dB.

I also can't hear the difference - and I'm also putting it down to (Motorhead induced) Tinnitus.

 

[ericw]

Listened on motherboard audio -> HD 600 headphones.

I found this pretty easy, the harmonic one had an "eee" tone vs the pure sine.

ABX result:

Spoiler

foo_abx 2.0.6d report
foobar2000 v1.4.1
2021-09-05 01:49:14

File A: sample1.flac
SHA1: c7fa3903f280279e07fd8a7981017b0634969711
File B: sample2.flac
SHA1: 6f45f66aef302ed42b9199074fbfaca1ab4dd708

Output:
DS : Primary Sound Driver
Crossfading: NO

01:49:14 : Test started.
01:49:28 : 01/01
01:49:35 : 02/02
01:49:42 : 03/03
01:49:46 : 04/04
01:49:50 : 05/05
01:49:57 : 06/06
01:50:04 : 07/07
01:50:11 : 08/08
01:50:15 : 09/09
01:50:20 : 10/10
01:50:25 : 11/11
01:50:31 : 12/12
01:50:35 : 13/13
01:50:40 : 14/14
01:50:46 : 15/15
01:50:50 : 16/16
01:50:50 : Test finished.

----------
Total: 16/16
p-value: 0 (0%)

-- signature --
68635c6d172cf4b65e6cc8021d1d7962d972821e

 

[pma]

Poll closed yesterday, my thanks to all who have participated. There were more people who could hear a difference (with ABX) than those who could not, 14 : 6, 70% : 30%. The test has confirmed that audibility of low order harmonic distortion highly depends on sound level, is good at low volume (detects <1% H2 distortion on a pure 1kHz tone) and that the ear ability to hear harmonic distortion decreases with higher volume.

 

[restorer-john]

and that the ear ability to hear harmonic distortion decreases with higher volume.

This is the key takeaway. It's ironic that audiophiles play music louder to hear so-called differences, when they actually should be listening at a lower level...

Thanks Pavel.

 

[xaviescacs]

This is the key takeaway. It's ironic that audiophiles play music louder to hear so-called differences, when they actually should be listening at a lower level...

Thanks Pavel.

Personally, as a newbie, the most important takeaway for me is that, although it was quite obvious, I've realized that typical musician habilities like hearing harmonics are also useful in this technical context.