[ColSaulTigh]

It's funny, because some amps don't seem to care at all about rectifiers, while others (like my Woo Audios) change dramatically by swapping out rectifiers.

I don't know enough about the subject to speak intelligently as to the "why" - maybe some of the elder statesmen of the tube world can chime in. @L0rdGwyn @Paladin79 @paradoxper and others are far better at explaining this than me.

 

[L0rdGwyn]

Well there are some things about tube vs. solid state that definitely will affect the sound. But there is also a very "tube voodoo" aspect of it that is hard to put your finger on.

The most significant different between using tube rectification and solid state rectification is the voltage drop across the diodes (silicon in solid state, tube diode as in tube rectifier). The drop across a silicon diode is usually on the order of 1V, whereas the average drop across a tube diode is around 30V, but it varies depending on the rectifier being used.

Why is this significant? It alters the operating point of the amplifying tube in the circuit, which will change their internal resistance, distortion characteristics, transconductance, etc. So, if swapping back and forth from solid state to tube rectification in the same circuit, you are likely causing drastic changes in in the tubes' operating points. Of course this could be accounted for by the designer such that there is additional voltage drop in the solid state power supply to compensate and ensure the tubes are always at the same operating points, but that is typically not done, it is usually accepted there will be a drop in the tube's operating voltage when using tube rectification.

Similarly, if you use different models of rectifier tubes, say a GZ34 vs. a 5U4, these tubes have different voltage drops, so again you are altering the tubes' operating points when you swap rectifiers, which will definitely alter the sound.

So what about swapping different makes of rectifiers of the same model, keeping the voltage drop identical and the tube operating points the same? That is where the tube voodoo comes in. People will question whether or not there truly are audible differences, of course many people would say yes, others no. As to why there are differences, it could be material differences in the rectifiers or perhaps the interaction of the audio signal with the rectifier in an unregulated power supply. But there is no certain answer that I have come across that explains it easily. It's similar to asking why some people prefer tubes over solid state amplifiers, that there is a "magic" to tube amplifiers.

Two other major technical differences in solid state vs. tube rectified supplies - solid state diodes can inject switching noise into the audio chain, but this can be accounted for using soft recovery diodes and proper solid state power supply design, so if designed appropriately it is a non issue. The other difference is power transformer ringing - a good power supply will snub any resonances in the power transformer, this ringing can reach the audio chain and alter the sound. Tube rectified supplies do this inherently - the high internal resistance of a tube rectifier will snub any ringing in a mains transformer. This is not the case with solid state diodes. As such, a proper snubber circuit should be applied in a well-designed solid state power supply. Unfortunately, this is very often skipped in commercial solid state amplifier supplies, as are the soft recovery diodes, current limiting resistance, and other elements of good solid state power supply design.

Anyway, that's the gist of where it stands in my mind. There are technical reasons for changes in sound, but there is some voodoo around tube rectification too, whether or not you believe it will be up to your ears.

 

[jonathan c]

Thank you, L0rdGwyn.

 

[ColSaulTigh]

Well there are some things about tube vs. solid state that definitely will affect the sound. But there is also a very "tube voodoo" aspect of it that is hard to put your finger on.

The most significant different between using tube rectification and solid state rectification is the voltage drop across the diodes (silicon in solid state, tube diode as in tube rectifier). The drop across a silicon diode is usually on the order of 1V, whereas the average drop across a tube diode is around 30V, but it varies depending on the rectifier being used.

Why is this significant? It alters the operating point of the amplifying tube in the circuit, which will change their internal resistance, distortion characteristics, transconductance, etc. So, if swapping back and forth from solid state to tube rectification in the same circuit, you are likely causing drastic changes in in the tubes' operating points. Of course this could be accounted for by the designer such that there is additional voltage drop in the solid state power supply to compensate and ensure the tubes are always at the same operating points, but that is typically not done, it is usually accepted there will be a drop in the tube's operating voltage when using tube rectification.

Similarly, if you use different models of rectifier tubes, say a GZ34 vs. a 5U4, these tubes have different voltage drops, so again you are altering the tubes' operating points when you swap rectifiers, which will definitely alter the sound.

So what about swapping different makes of rectifiers of the same model, keeping the voltage drop identical and the tube operating points the same? That is where the tube voodoo comes in. People will question whether or not there truly are audible differences, of course many people would say yes, others no. As to why there are differences, it could be material differences in the rectifiers or perhaps the interaction of the audio signal with the rectifier in an unregulated power supply. But there is no certain answer that I have come across that explains it easily. It's similar to asking why some people prefer tubes over solid state amplifiers, that there is a "magic" to tube amplifiers.

Two other major technical differences in solid state vs. tube rectified supplies - solid state diodes can inject switching noise into the audio chain, but this can be accounted for using soft recovery diodes and proper solid state power supply design, so if designed appropriately it is a non issue. The other difference is power transformer ringing - a good power supply will snub any resonances in the power transformer, this ringing can reach the audio chain and alter the sound. Tube rectified supplies do this inherently - the high internal resistance of a tube rectifier will snub any ringing in a mains transformer. This is not the case with solid state diodes. As such, a proper snubber circuit should be applied in a well-designed solid state power supply. Unfortunately, this is very often skipped in commercial solid state amplifier supplies, as are the soft recovery diodes, current limiting resistance, and other elements of good solid state power supply design.

Anyway, that's the gist of where it stands in my mind. There are technical reasons for changes in sound, but there is some voodoo around tube rectification too, whether or not you believe it will be up to your ears.

See, I knew someone in here could do a much better job explaining the nitty-gritty! Thank you for today's lesson. Tomorrow's subject will be "Macro-Economics and the Driving Forces Behind Inflation".

Volunteers?

 

[L0rdGwyn]

Thank you, L0rdGwyn.

No problem, I'm always up for a good tube rant.

 

[paradoxper]

It's funny, because some amps don't seem to care at all about rectifiers, while others (like my Woo Audios) change dramatically by swapping out rectifiers.

I don't know enough about the subject to speak intelligently as to the "why" - maybe some of the elder statesmen of the tube world can chime in. @L0rdGwyn @Paladin79 @paradoxper and others are far better at explaining this than me.

Audiophiles exaggerate the minute. To a strong degree. To often delusioned measures.

If you ask xyz designer, they'll prudently share rectifiers won't matter, and they shouldn't, not to the nth degree.

They also don't exhaust rolling such as the enthralled audiophile.

If you can't articulate the why in relation to the technical how, you shouldn't worry.
Maximize your subjectivity and personal anecdote with some measure of restraint.

It is the world of audio, including a world of the woo.

 

[L0rdGwyn]

The number one reason to use a tube rectifier is because... they're cool but in basically every technical sense, solid state diodes are higher performing. Any technical advantage a tube rectifier might have (i.e., transformer snubbing, lack of switching noise, soft start characteristics) can be accounted for in a solid state supply.

 

[ColSaulTigh]

The number one reason to use a tube rectifier is because... they're cool but in basically every technical sense, solid state diodes are higher performing. Any technical advantage a tube rectifier might have (i.e., transformer snubbing, lack of switching noise, soft start characteristics) can be accounted for in a solid state supply.

So if I could ask a follow-up to this - what causes "tube sag", something that I immediately picked up on when I got my first Woo Audio WA6-SE? Changing the rectifier had an immediate effect on the sound, removing that "out of breath" characteristic. Is that just quality of rectifier, or something else entirely?

 

[protoss]

tube sag

Microphonic, air displacement, compression, decay and bloom. Somewhat in that order causes slowness and all done with voltage increase or decrease.

 

[L0rdGwyn]

So if I could ask a follow-up to this - what causes "tube sag", something that I immediately picked up on when I got my first Woo Audio WA6-SE? Changing the rectifier had an immediate effect on the sound, removing that "out of breath" characteristic. Is that just quality of rectifier, or something else entirely?

Tube sag doesn't really exist in these types of amplifiers, SET amplifiers are single-ended class A, meaning the DC current draw on the power supply is essentially constant, which means there is a constant voltage drop across the resistance of the power supply. By Ohm's law (V=I*R), a current going through a finite resistance will result in a known voltage drop. If the current increases, the voltage drop will increase proportionately. BUT because we are talking class A, the current draw is constant and is defined by the DC bias points of the tubes. Since the current is constant, that means the voltage drop across the rectifier and power supply is constant.

Now, if you look at a class AB push-pull amplifier, once the amplifier enters class B, the current is no longer constant! The amplifier circuit now represents a dynamic load, the current drawn from the power supply will be variable.

What does that mean in terms of the voltage drop across the power supply? If the current is changing, then the voltage must also be changing, so the bias points of the tubes will be changing. THIS is sag, the voltage of the power supply will "sag" (i.e., go down) as the amplifier draws more current, which is undesirable. This is something that needs to be accounted for in designs that have a variable DC current load.

For single-ended class A tube amps, like headphone SET amps, sag doesn't occur since the current draw is constant.

In terms of what you're hearing when you change rectifiers, it could be as I described in my other post, that the bias points are being altered when you change rectifiers. A higher voltage drop rectifier might sound "slow" compared to a lower voltage drop rectifier as the tubes are at a colder bias point (i.e., lower plate voltage, lower bias current), which means higher internal resistance, which means a slightly higher output impedance and likely higher distortion. If you are using two rectifiers of the same type that are measurably identical in terms of their voltage drops, then I have no idea, and I will call that voodoo. Maybe material differences are causing an appreciable change in sound, maybe there is an element of placebo effect, I can't say for sure.

Microphonic, air displacement, compression, decay and bloom. Somewhat in that order causes slowness.

 

[jonathan c]

No problem, I'm always up for a good tube rant.

Do you normally rant in 5v or in 4v? How great is the voltage drop after you post?…

 

[ThanatosVI]

Tube sag doesn't really exist in these types of amplifiers, SET amplifiers are single-ended class A, meaning the DC current draw on the power supply is essentially constant, which means there is a constant voltage drop across the resistance of the power supply. By Ohm's law (V=I*R), a current going through a finite resistance will result in a known voltage drop. If the current increases, the voltage drop will increase proportionately. BUT because we are talking class A, the current draw is constant and is defined by the DC bias points of the tubes. Since the current is constant, that means the voltage drop across the rectifier and power supply is constant.

Now, if you look at a class AB push-pull amplifier, once the amplifier enters class B, the current is no longer constant! The amplifier circuit now represents a dynamic load, the current drawn from the power supply will be variable.

What does that mean in terms of the voltage drop across the power supply? If the current is changing, then the voltage must also be changing, so the bias points of the tubes will be changing. THIS is sag, the voltage of the power supply will "sag" (i.e., go down) as the amplifier draws more current, which is undesirable. This is something that needs to be accounted for in designs that have a variable DC current load.

For single-ended class A tube amps, like headphone SET amps, sag doesn't occur since the current draw is constant.

In terms of what you're hearing when you change rectifiers, it could be as I described in my other post, that the bias points are being altered when you change rectifiers. A higher voltage drop rectifier might sound "slow" compared to a lower voltage drop rectifier as the tubes are at a colder bias point (i.e., lower plate voltage, lower bias current), which means higher internal resistance, which means a slightly higher output impedance and likely higher distortion. If you are using two rectifiers of the same type that are measurably identical in terms of their voltage drops, then I have no idea, and I will call that voodoo. Maybe material differences are causing an appreciable change in sound, maybe there is an element of placebo effect, I can't say for sure.

Thx for those explanations. Interesting to read

 

[jonathan c]

See, I knew someone in here could do a much better job explaining the nitty-gritty! Thank you for today's lesson. Tomorrow's subject will be "Macro-Economics and the Driving Forces Behind Inflation".

Volunteers?

Preview: demand pull, supply push, excess monetary growth, expectations.

 

[L0rdGwyn]

Thx for those explanations. Interesting to read

No problem. I should probably clarify that sag is undesirable in a hifi amplifier, but may be desirable in a guitar amplifier. But I am no expert in guitar amplifier design, not something I've read into deeply because... I don't play guitar

 

[VandyMan]

Maybe material differences are causing an appreciable change in sound, maybe there is an element of placebo effect, I can't say for sure.

Placebo is always a possibility, but I experienced the same thing with my Woo WA6SE (2nd gen). Not proof by any means, but another data point.