I was wondering if anyone had read any technical papers on the source of various tube-based sonic effects.
Let's take into consideration the imaging effect most commonly referred to as 'air'. It it caused by transparency on the part if the tube or is it caused be an interplay of odd-order harmonics within the tube? Or perhaps something else?
Let's take into consideration the imaging effect most commonly referred to as 'air'. It it caused by transparency on the part if the tube or is it caused be an interplay of odd-order harmonics within the tube? Or perhaps something else?
There are lots of various "theories" (mostly audiophool old wives' tales) but very little that's in any way definitive. You don't see anything concerning this in something like the Radiotron Designers Handbook. A big part of the problem lies with the terminology and limitations of language. WTH does "air" mean anyway? It's as difficult to describe a sound as it is to tell a man born blind what "blue" looks like.
You can measure distortion spectra and draw some general conclusions, but that's about the best you can do. You don't see the subject even mentioned in EE classes at all. Which is probably why a good deal of your Big Box equipment sounds as horrible as it does.
This was a big problem during the development of my latest project: virtually nothing concerning the sonic qualities of a cascoded LTP. All that was left was to actually put one together and hope for the best. Forch, it worked out just great. Different VTs will have different sonic "personalities" depending on whether they produce mainly the sonically benign h2 and h3, or whether they produce lots of dissonant high order harmonics. About all you can do is design and listen. Then, and only then, can you decide what needs to be done so far as NFB correction is concerned. Keep in mind that what sounds good to you may not sound so good to some other listener.
You can measure distortion spectra and draw some general conclusions, but that's about the best you can do. You don't see the subject even mentioned in EE classes at all. Which is probably why a good deal of your Big Box equipment sounds as horrible as it does.
This was a big problem during the development of my latest project: virtually nothing concerning the sonic qualities of a cascoded LTP. All that was left was to actually put one together and hope for the best. Forch, it worked out just great. Different VTs will have different sonic "personalities" depending on whether they produce mainly the sonically benign h2 and h3, or whether they produce lots of dissonant high order harmonics. About all you can do is design and listen. Then, and only then, can you decide what needs to be done so far as NFB correction is concerned. Keep in mind that what sounds good to you may not sound so good to some other listener.
Well, tubes are made of glass, and glass is transparent so obviously the sound they make is transparent too. Solid state amplifiers are made out of silicon, which is sand. Sand is gritty, so that's why all solid state amplifiers have a gritty sound. But air? Everyone knows that vacuum tubes don't have any air in them. That's why they're called vacuum tubes. I can't imagine how they could sound "airy".
[/sarcasm]
Just kidding.
[/sarcasm]
Just kidding.
Come to think of it, you're right. The tubes which sound gritty to me often have thick glass and it's the presence of all that sand that does it... interesting...
So what we need is tubes made from thin lead crystal or zircon to sound good. There goes my dream of making tubes from beer bottles!!! lol
set BS/off
I'm going with the transparency theory...
This all started with me getting some '52 Soviet 6C2C tubes for my handmade amp. The things are incredible! It would be great if Russia was still producing tubes of that quality..
So what we need is tubes made from thin lead crystal or zircon to sound good. There goes my dream of making tubes from beer bottles!!! lol
set BS/off
I'm going with the transparency theory...
This all started with me getting some '52 Soviet 6C2C tubes for my handmade amp. The things are incredible! It would be great if Russia was still producing tubes of that quality..
Lynn Olson wrote the following on THD -
http://www.nutshellhifi.com/library/FindingCG.html
worth a further discussion?
http://www.nutshellhifi.com/library/FindingCG.html
worth a further discussion?
SY said:
The best example of 'air' that I can come up with is 6:05 into 'Andalu' by Chris Spheeris. It's the effect where the sound stage becomes huge and even the most subtle of sounds are caught in echos. Of course the example I cite is more than just 'air'. It's brilliant!!
TubeHead Johnny said:
Various 6J5s in the amp I am using sound quite different...
"Air" in certain tubes
I think it is the tube design more than anything else.
In my SE amp I've tried various tubes and
the KT66s have the most air by far.
All the EL34s, KT77s, 6550s and KT88s that
I have tried never had the air of KT66s.
They were all in triode with the same OPT.
Saratov seems to have the most because
it is lacking in the lower regions.
The Genelex re-issue has a balanced sound
and still retains the air quality to it.
I do love the air on certain recordings because
it carries you to an ethereal place.
Sarah McLachlan Surfacing Album is great
to listen to with the proper air factor.
DanL
I think it is the tube design more than anything else.
In my SE amp I've tried various tubes and
the KT66s have the most air by far.
All the EL34s, KT77s, 6550s and KT88s that
I have tried never had the air of KT66s.
They were all in triode with the same OPT.
Saratov seems to have the most because
it is lacking in the lower regions.
The Genelex re-issue has a balanced sound
and still retains the air quality to it.
I do love the air on certain recordings because
it carries you to an ethereal place.
Sarah McLachlan Surfacing Album is great
to listen to with the proper air factor.
DanL
I get the most air after broccoli, which is pretty much what I think of discussions of "airiness" in tubes.
If you go changing tube types in amplifiers without changing the conditions they operate under then you are going to get changes in the sound, that's pretty obvious. But categorising these changes, which are simply due to a new tube generating more or less distortion when operating under the same conditions as another, "air" (or lack thereof) doesn't seem to have a great foundation in any science I've heard of.
Sorry, I just get grumpy with these terms because they are purely personal, subjective and entirely dependent on the piece of gear in front of you. Someone on the other side of the world with a different amp is probably not going to discover the same KT66 air factor as you, nor could they.
</rant>
If you go changing tube types in amplifiers without changing the conditions they operate under then you are going to get changes in the sound, that's pretty obvious. But categorising these changes, which are simply due to a new tube generating more or less distortion when operating under the same conditions as another, "air" (or lack thereof) doesn't seem to have a great foundation in any science I've heard of.
Sorry, I just get grumpy with these terms because they are purely personal, subjective and entirely dependent on the piece of gear in front of you. Someone on the other side of the world with a different amp is probably not going to discover the same KT66 air factor as you, nor could they.
</rant>
"Air" in tubes
I stated that I had a SE trioded amp and
I will also state that I use no feedback.
See full schematic here:
http://www.integracoustics.com/images/DanL/SE KT66.GIF
I tried various cathode resistors on various tubes
to try to get the best sound but the same OPT.
That rules out OPT frequency response dependance.
Also the speaker must be extended enough to reproduce
the top octave well to get the air.
My tweeters are Dynaudio D-260s with GE Motor Run Caps
which reproduce highs clearly and have great extension.
I don't see the problem of having air or airy highs
as personal or subjective, it's pretty obvious whether
the system can produce airy highs or it won't.
I also have a SS Marantz SR-66 reciever that has great
air too - that's why I kept it for my other system.
If someone's system is capable of having air
then it can with the proper tube.
If it doesn't then you need to find the weakness
in the system or live without it.
Me - I like plenty of air in my highs.
DanL
I stated that I had a SE trioded amp and
I will also state that I use no feedback.
See full schematic here:
http://www.integracoustics.com/images/DanL/SE KT66.GIF
I tried various cathode resistors on various tubes
to try to get the best sound but the same OPT.
That rules out OPT frequency response dependance.
Also the speaker must be extended enough to reproduce
the top octave well to get the air.
My tweeters are Dynaudio D-260s with GE Motor Run Caps
which reproduce highs clearly and have great extension.
I don't see the problem of having air or airy highs
as personal or subjective, it's pretty obvious whether
the system can produce airy highs or it won't.
I also have a SS Marantz SR-66 reciever that has great
air too - that's why I kept it for my other system.
If someone's system is capable of having air
then it can with the proper tube.
If it doesn't then you need to find the weakness
in the system or live without it.
Me - I like plenty of air in my highs.
DanL
Boston Bean factor
From the description(s) of "air" I'll take a guess at what is happening. The amplifier has characteristic curvature in the low signal amplitude region, that produces 2nd harmonic distortion for the typically low amplitude high frequencies. The typically larger amplitude bass frequencies may or may not have the noticeable 2nd harmonic effect, depending on the larger full characteristic curvature.
Don
From the description(s) of "air" I'll take a guess at what is happening. The amplifier has characteristic curvature in the low signal amplitude region, that produces 2nd harmonic distortion for the typically low amplitude high frequencies. The typically larger amplitude bass frequencies may or may not have the noticeable 2nd harmonic effect, depending on the larger full characteristic curvature.
Don
I have gas....
To hear the effect about which I speak, have a listen to my reference music that I often use to compare the ability of various systems. Not only is the program of this music superb but so is the recording.
http://www.imeem.com/people/-7hsqJ/music/O0ONDv6Y/andalu/
The recorded effects to which I refer start at 6:04. I have heard systems which only botch this and systems that do such an incredible job that one realizes it's amazing what a person can hear...
To hear the effect about which I speak, have a listen to my reference music that I often use to compare the ability of various systems. Not only is the program of this music superb but so is the recording.
http://www.imeem.com/people/-7hsqJ/music/O0ONDv6Y/andalu/
The recorded effects to which I refer start at 6:04. I have heard systems which only botch this and systems that do such an incredible job that one realizes it's amazing what a person can hear...
Re: I have gas....
OK, I see what you're getting at here. This is a function of linearity, and having minimal IMD. IMD is worse than harmonic distortion, especially where signal frequencies occur close together.
You never showed us a schemo, but if that's the case, it doesn't include any NFB, does it? With low gain devices like VTs, the circuit performance depends more heavily on the active component than it does with high gain devices like transistors.
As I don't have the inclination to get samples of each of the bazillion makes and models of the various VTs I use, I prefer to add some NFB to make circuit performance more independent of active and passive components.
Rodango said:
OK, I see what you're getting at here. This is a function of linearity, and having minimal IMD. IMD is worse than harmonic distortion, especially where signal frequencies occur close together.
You never showed us a schemo, but if that's the case, it doesn't include any NFB, does it? With low gain devices like VTs, the circuit performance depends more heavily on the active component than it does with high gain devices like transistors.
As I don't have the inclination to get samples of each of the bazillion makes and models of the various VTs I use, I prefer to add some NFB to make circuit performance more independent of active and passive components.
Well the ear canal causes significant harmonic distortion due to its shape. However this distortion, which mainly is low order 2nd and 3rd harmonic, is corrected by the brain and CNS to get the perception of a pure sound.
This is why tube amps, especially SETs, sound quite good- the low order distortion created is not even heard, even at several percent THD; as it is quite similar to the distortion spectrum of the ear canal; and therefore gets corrected by the brain.
However there are other psychoacoustic effects- 2nd harmonic distortion has the effect of making the bass appear boomy and rich. I'm not sure what 3rd and 5th give, but considering they are the major factors of PP amps, maybe they add more bite and more imaging...
This is why tube amps, especially SETs, sound quite good- the low order distortion created is not even heard, even at several percent THD; as it is quite similar to the distortion spectrum of the ear canal; and therefore gets corrected by the brain.
However there are other psychoacoustic effects- 2nd harmonic distortion has the effect of making the bass appear boomy and rich. I'm not sure what 3rd and 5th give, but considering they are the major factors of PP amps, maybe they add more bite and more imaging...
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.