TUBE AMPS AND DISTORTION

[sofrep811]

Just a question: Is the distortion a creation of the tubes or the transformer? If transformer, what are tubes offering to the audio as it is expelled?

Same goes with tubes and DAC--what exactly are tubes adding to help a Digital Audio Converter?

Some very intelligent folks here and figured someone can advise.

 

[NiagaraPete]

Same goes with tubes and DAC--what exactly are tubes adding to help a Digital Audio Converter?

Nothing. Why would you want distortion in signal processing?

 

[DVDdoug]

Is the distortion a creation of the tubes or the transformer?

Tube amps don't necessarily have audible distortion except for guitar amp's which are supposed to sound good when overdriven.

I think it's both. From what I understand tubes tend to soft clip when overdriven whereas solid state transistors & MOSFETs tend to hard clip, but it depends on the circuit design.

 

[levimax]

Distortion is caused by both tubes and transformers. A good tube amp can have inaudible distortion especially older designs that strived to be "Hi-Fi". Unfortunately many modern tube designs are poor performers due to cost cutting, poor designs, or even intentionally designing them to have higher distortion. While cool to look at tube amps don't / can't improve sound quality... good ones do no harm, poor ones add distortion which hurts sound quality.

 

[GXAlan]

What are tubes offering to the audio as it is expelled?
....
Some very intelligent folks here and figured someone can advise.

It's the same question as "what are transistors offering to the audio as it is expelled?"

You can have transistors that measure well/don't measure well the same way you can have tubes that measure/don't measure well.

If you look at Stereo Magazine, they have the 50mW distortion on a tube amp beating that of even the Benchmark AHB2. Obviously as you go up in power the tides turn.

I think the most precise statement is that tubes sound different. They're not better. But some people prefer the difference. The even vs. odd harmonics is more a reflection of the topology/design rather than transistor vs. tube.

I would probably say that microphonics are content specific, room and volume specific modifications to the sound. The same way you might like Auro3D or other upmixing, it's possible that the fatter sound from the extra reverberation adds to the change in tone that tube owners like. We can see how digital dither can improve things. You are adding technically noise or extraneous data that is not part of the original recording and yet it makes measurements better. https://www.audiosciencereview.com/...ds/denon-da-500-vintage-r2r-dac-review.17232/

 

[DonH56]

It's (very) complicated and depends upon the circuit as well as the components. Tubes intrinsically have a distortion series that is factorial, while bipolar transistors' distortion series is exponential. Fundamentally that means tubes have lower distortion than bipolar transistors. But transistor circuits are often differential so they suppress even-order distortion terms, and bipolar transistors have higher gain so you can apply more feedback to further lower distortion. Transformers can saturate and have hysteresis that increases their distortion particularly at lower frequencies, and since they are big coils can affect high-frequency response. There are so many details and nuances in circuit design that it would take a big (several, probably) book and/or some coursework to really explain the differences, pro and con. These few lines aren't even a start...

A good tube circuit can have inaudible distortion, but many designs seem to intentionally emphasize (and market) distortion to provide what they perceive people expect as "tube sound".

IMO - Don

 

[DonR]

It's (very) complicated and depends upon the circuit as well as the components. Tubes intrinsically have a distortion series that is factorial, while bipolar transistors' distortion series is exponential. Fundamentally that means tubes have lower distortion than bipolar transistors. But transistor circuits are often differential so they suppress even-order distortion terms, and bipolar transistors have higher gain so you can apply more feedback to further lower distortion. Transformers can saturate and have hysteresis that increases their distortion particularly at lower frequencies, and since they are big coils can affect high-frequency response. There are so many details and nuances in circuit design that it would take a big (several, probably) book and/or some coursework to really explain the differences, pro and con. These few lines aren't even a start...

A good tube circuit can have inaudible distortion, but many designs seem to intentionally emphasize (and market) distortion to provide what they perceive people expect as "tube sound".

IMO - Don

^ this.... so many times, over and over... this.

 

[DMill]

A good tube circuit can have inaudible distortion, but many designs seem to intentionally emphasize (and market) distortion to provide what they perceive people expect as "tube sound".

It’s interesting that I comment more on my tube amp setup vs my solid state but here it goes… I don’t hear distortion at low SPL. It sounds good and looks amazing. The neighbors wives ask me to reco good music. So I leave the tubes out for dinner parties. My other system is a tiny bit better. But just a tiny bit. Just saying.

 

[sofrep811]

Many thanks. All. I've run a single ended A (Cayin) for a few years and loved it. I don't recall any distortion in it. Seemed smooth as silk.

 

[egellings]

Distortion in tube amps or anything else is a result of the circuit design, mostly.

 

[Robin L]

Speaking from experience, said experience coming from a number of vintage amps and a headphone amp that was direct coupled to triodes in the output stage, transformers are producing a lot of that "tube sound". The headphone amp was a Stax energizer/amp, the power amps were from Dynaco, Fisher, Marantz and Scott. The Scott amp was the most recent design and had the least fogginess of the batch. I would suspect modern amps would close the gap, but older tube designs had a sound quite different from transistor designs. I suspect that SETs have more of that "tube sound", at least based on what little exposure I've had with SETs. A lot of ideas about tube amps come from the sound of designs more than 60 years old.

 

[DSJR]

I still have some subjectivism left, but believe that when I DO hear repeatable differences, there has to be a measurable reason why!

I started to type another mini-essay, but then remembered where I am All I can say is that where line stages are concerned, I don't think it matters whether it's valve, solid state discrete or a bloomin' op-amp as long as it's been designed carefully, as any of these should be transparent and as noise free as needed. Power amps are different as some circuits are speaker-immune and others patently NOT, altering their frequency response with a typical speaker loading.

These days, I like to keep variables down, so would select source and amps to be as honest, neutral and 'safe' as possible as long as they're transparent to the recording/mastering intentions of the music, leaving the visuals and cost-fi* aspects out these days. I've been down the valve route several times in my life and despite treasuring and curating a set of sensitively rebuilt Quad II's which only come out to play every ten years or so, I don't need that any more and can't afford it anyway (no disrerespect to the valve lovers here whom I deeply respect).

*The more expensive it is, the better it's often thought to be!

 

[levimax]

Speaking from experience, said experience coming from a number of vintage amps and a headphone amp that was direct coupled to triodes in the output stage, transformers are producing a lot of that "tube sound". The headphone amp was a Stax energizer/amp, the power amps were from Dynaco, Fisher, Marantz and Scott. The Scott amp was the most recent design and had the least fogginess of the batch. I would suspect modern amps would close the gap, but older tube designs had a sound quite different from transistor designs. I suspect that SETs have more of that "tube sound", at least based on what little exposure I've had with SETs. A lot of ideas about tube amps come from the sound of designs more than 60 years old.

Have you ever done a level matched double blind test with amps? I tried it with a Dynaco ST 70 vs a Neurochrome MOD 86 and was unable to tell any difference. Of course I may be deaf but I also had a 16 year old try it and he could tell no difference either. I can tell you that while setting up the test (which was not easy to get quiet seamless switching and allowing for the tube amp not going open circuit) the differences were pretty clear to me until the final "level matching" at which point all differences disappeared. I am not saying there were no differences as I could measure some differences in FR with REW due to the internal resistance of the tube amp (fraction of a dB) but I could not hear it. For me at least there is no such thing as "tube sound" for a well designed tube amp.

 

[Robin L]

Have you ever done a level matched double blind test with amps? I tried it with a Dynaco ST 70 vs a Neurochrome MOD 86 and was unable to tell any difference. Of course I may be deaf but I also had a 16 year old try it and he could tell no difference either. I can tell you that while setting up the test (which was not easy to get quiet seamless switching and allowing for the tube amp not going open circuit) the differences were pretty clear to me until the final "level matching" at which point all differences disappeared. I am not saying there were no differences as I could measure some differences in FR with REW due to the internal resistance of the tube amp (fraction of a dB) but I could not hear it. For me at least there is no such thing as "tube sound" for a well designed tube amp.

I have never done a level matched double-blind test. And there have might have been additional variables I have not considered. I did run a recapped Marantz 8B as the amp for my Stax earspeakers using a Stax energizer and compared that to using the OTL Stax Tube amp/energizer. The difference was sufficiently pronounced as to not require a DBT. With the 8B it seemed that the frequency extremes were out of focus, only the midrange in focus, whereas the Stax energizer/amp was focused in the frequency extremes as well as the midrange. Note that I'm not saying the 8B is necessarily a good or bad design, and I'm not ruling out the possibility that the Marantz amp was not properly aligned. However, I suspect that a lot of the so-called "Myths" of "tube sound" has to do with older designs that don't reach the levels of performance of modern gear, have worn out parts, etc. It should be noted that the Stax earspeakers I had could be accused of an excess of detail, perhaps a FR deviation that emphasizes treble detail. In any case, that's what happened and as I said previously, the difference in overall SQ was sufficiently marked as to not require a DBT.

 

[Killingbeans]

Tube amps don't necessarily have audible distortion except for guitar amp's which are supposed to sound good when overdriven.

It made me think of this thread that flew under the radar a few days ago:

Interesting video on guitar amp tone.

I found this video on a musician trying to find out what makes up a guitar amp's tone interesting and applicable to ASR.

www.audiosciencereview.com

Not really sure whether it's applicable to the discussion at hand in any way.

 

[Sokel]

I listen to AR REF160M all the time (driving big horns)
It's really really hard to perceive what you hear as distortion.Their site does not even state numbers about it.
There is something there but I'm convinced it's 100% deliberate,it's like a recipe.
Sometimes I try hard to find something missing or altered,it's impossible.
I'm sure I would not distinguish it in a AB blind but not for the right reasons (for good or bad)

 

[egellings]

I have never done a level matched double-blind test. And there have might have been additional variables I have not considered. I did run a recapped Marantz 8B as the amp for my Stax earspeakers using a Stax energizer and compared that to using the OTL Stax Tube amp/energizer. The difference was sufficiently pronounced as to not require a DBT. With the 8B it seemed that the frequency extremes were out of focus, only the midrange in focus, whereas the Stax energizer/amp was focused in the frequency extremes as well as the midrange. Note that I'm not saying the 8B is necessarily a good or bad design, and I'm not ruling out the possibility that the Marantz amp was not properly aligned. However, I suspect that a lot of the so-called "Myths" of "tube sound" has to do with older designs that don't reach the levels of performance of modern gear, have worn out parts, etc. It should be noted that the Stax earspeakers I had could be accused of an excess of detail, perhaps a FR deviation that emphasizes treble detail. In any case, that's what happened and as I said previously, the difference in overall SQ was sufficiently marked as to not require a DBT.