• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Rectifier "Sound"?

OK, so this is going to be a bit of a loaded question and just might open a big can of worms but I really want to try to stay open-minded on this before passing judgement.

I was just reading a huge review of rectifier tubes over at a popular headphone-listeners forum and I have to say I am skeptical.

I'm a tube newb, but as I understand it, a rectifier is just what it sounds like; it converts AC to DC which is then filtered. The result is the B+. The audio signal does not flow through the tube at all.

I do understand that different rectifier types work and behave differently. A classic example that I have seen many times (typically on guitar amp forums) is that a 5U4 tube will cause more "sag" vs. a 5AR4 tube and in some cases this is desirable. Either way, changing the rectifier type can change the character, quality, or "speed"? of the rectification in such a way that it is audible.

However, I find it doubtful that a specific 5U4 tube (as compared to other 5U4 tubes) can provide "Hauntingly beautiful tone, lots of air, great layering, transparency above the rest, and just perfect extension in both the highs and lows, deep bass, palpable, lucid midrange, holographic 3D staging and imaging ... amazing! Overall balance is great as its neither overly warm nor bright, transients and PRaT are as good as they get with tubes (like seriously, beautiful liquid transients!)"

Again, the audio signal isn't passing through this tube. Its job is to convert AC to DC. Am I crazy for thinking the above is 100% bollocks? I wonder what DF96 would have said.
 
Last edited:
Several times I think that those who say that an amplifier changes its sound changing a rectifier, a speaker cable or a coupling cap are, generally speaking, lying.

A cap for example, is a complete two terminal network at audio frequencies and a four terminal at very high frequencies. How a guy can distinguish some pF more than others is for me, in-understandable.

Now, may be a rectifier, whose internal impedance varies from one type to other, is reasonable that the sound at the speaker changes. How much, I can't be able to say.
 
Hi cogitech
This is one I also thought made no sense. I cannot explain it, but choice of rectifier tube can make a huge difference on sound. I use them on the output stage of a triode buffered DAC. I started with Russian diodes. One failed after a year, and I bought NOS 5U4G tubes (Marconi). OMG! the sound was much richer and the bass response much more substantial. I think it was the bass and mid bass that benefited the most. I am still surprised by the impact.

Next, I am going to try the "perfect rectifier" GB to replace solid state rectifiers. It uses a controller and mosfets to replace a bridge rectifier.
 
It does. Where do you think the electrons flowing through the louspeaker come from? The reservoir capacitor is just part of the story.

Well, not really. That's not the way I understand it.

The electrons from the rectifier flow through the power tubes and thereby amplify the audio signal (and out to the loudspeaker). The audio signal does not actually go anywhere near the rectifier.

Correct me if I am wrong.
 
Now, may be a rectifier, whose internal impedance varies from one type to other, is reasonable that the sound at the speaker changes. How much, I can't be able to say.

Certainly, if the difference in tubes is significant enough, then we will hear it. The guitar amp example makes sense to me. The 5U4 isn't as quick to provide power (and also drops the voltage considerably more) compared to 5AR4. It seems obvious that this will be audible.

However, when comparing 2 5AR4 tubes with extremely similar specifications and to claim such dramatic superiority of one over the other... I dunno.
 
Member
Joined 2011
Paid Member
In a hi-fi amp, the power supply should be as stiff as they come IMHO, and changing an HER308 to a UF5408 I don't notice any change at all. Tube rectifiers have a certain sag to them, this is why powerful push pull amps from the old days had choke input. Gives you better regulation.

Fast forward to now. Unless you purposefully WANT you rectifier to affect the sound such as in a guitar amp, you SHOULDN'T notice a difference. As Ling said, if it's designed well in the first place, it should work anyway.
 
Actually DC flows though the rectifier, audio signal though the nearest capacitor available.

I suppose I don't understand how a tube rectifier works then.

To me, a rectifier is a device where AC enters one side and DC comes out the other side, like in a typical full-wave bridge:

An externally hosted image should be here but it was not working when we last tested it.
 
Properly speaking, it is not a question of speed. I ran a 35W4 at 100KHz in a tube SMPSU and rectified -350V as the better SS diode. The problem is the distance from the cathode to the anode that in a 5U4 is larger and more irregular (Because of the inverted V form of its heater) the the 5AR4 rectifier, with less distance and cylindrical symmetry.
 
In a hi-fi amp, the power supply should be as stiff as they come IMHO, and changing an HER308 to a UF5408 I don't notice any change at all. Tube rectifiers have a certain sag to them, this is why powerful push pull amps from the old days had choke input. Gives you better regulation.

Fast forward to now. Unless you purposefully WANT you rectifier to affect the sound such as in a guitar amp, you SHOULDN'T notice a difference. As Ling said, if it's designed well in the first place, it should work anyway.

This is the way I understand it as well. You've said it much more clearly.
 
Member
Joined 2011
Paid Member
Well, not really. That's not the way I understand it.

The electrons from the rectifier flow through the power tubes and thereby amplify the audio signal (and out to the loudspeaker). The audio signal does not actually go anywhere near the rectifier.

Correct me if I am wrong.

Test it. Connect a small (0.01uf) cap to the power supply and a scope to see the AC ripple. Blast your music and tell me you don't see it on the scope.
 
Test it. Connect a small (0.01uf) cap to the power supply and a scope to see the AC ripple. Blast your music and tell me you don't see it on the scope.

I don't have an O-scope to test with.

But wait a second, are you saying that the AC audio signal (input) is sent through the rectifier tube and converted to DC, then on through the driver and power tubes?

If that's correct, then I really need to re-learn everything I know (which isn't much, admittedly).
 
Hi cogitech
This is one I also thought made no sense. I cannot explain it, but choice of rectifier tube can make a huge difference on sound. I use them on the output stage of a triode buffered DAC. I started with Russian diodes. One failed after a year, and I bought NOS 5U4G tubes (Marconi). OMG! the sound was much richer and the bass response much more substantial. I think it was the bass and mid bass that benefited the most. I am still surprised by the impact.

Next, I am going to try the "perfect rectifier" GB to replace solid state rectifiers. It uses a controller and mosfets to replace a bridge rectifier.

Sorry. Not ignoring you. I find your perspective interesting but I remain skeptical. I am in the process of building a SET and will be trying various rectifiers (including SS). It will be interesting to see if I can hear the difference.
 
Aiken amps has a good article on this, albeit in the realm of guitar rather than hifi.

What is "Sag"?

Typically, tube rectifiers have an inherent impedance that is several orders of magnitudes greater than a solid state diode. So the power supply voltage will generally drop more in response to increasing current demand with tube rectifiers than solid state. I am happy to speculate that this effect is audible. In the context of a hifi amplifier, especially non class A, my thought is that this voltage sag as function of signal amplitude likely manifests as measurable distortion. Whether it sounds 'good' on a hifi amp is probably a subjective distinction to make and dependent on whether the effect manifests as even or odd order harmonics with respect to the fundamental frequency of a test (or music) signal. Some math could probably bring us that answer.

Philosophically (although I am certainly no expert), I agree with what's been posted already; that tube rectifiers and their inherent output impedance makes them generally inferior to a solid state diode. However, as with any diode (SS or otherwise) the current / voltage curves are non-linear so depending on the particular application, there could be a convenient piece-wise voltage drop in the current range demanded, but that convenience may only occur when a PT has a bit too high of a secondary voltage.

Finally, solid state rectifier diodes often times support current or reverse bias voltage ratings that are beyond a garden variety rectifier tube.
 
Member
Joined 2011
Paid Member
No... I'm saying the AC of the signal will be imposed on the DC coming from the power supply... Not so much in a class A design, but for sure in an AB/B amp. The signal causes a current which causes a voltage sag that corresponds... It's not uncommon to have an older cheap low powered amp that has a change of 20 volts or more from silent output to full blast. The capacitors try and decouple that signal, but there's still traces of "music" on the power supply because of it.
 
Good afternoon from Greece! I think tube rectifiers is not good. First these devices failed so easy,especialy new production,the nos is the different story but it is so expensive.Second the internal resistance becomes bigger slowly when the tube degraded ,lower voltage in circuits,bad sound as result.Third tube rectifiers not like big capacitance,ripple as result.Fourth not stable quality in sound different tube,different internal resistance,different sound! Five energy wasted device needs filament extra windings more expensive transformers,and extra heat! The best solution is ss rectification no extra power, lower internal resistance,no problem in filter capacitance,and finally better sound i think more dynamic. In early days 50-60,the tube amps use tube rectifiers only,not reliably ss devices like selenium rectifiers. In present time you can buy cheap,and very reliably ss devices,but some of us likes tube rectifiers,fot more mellow and relaxed sound.
 
No... I'm saying the AC of the signal will be imposed on the DC coming from the power supply... Not so much in a class A design, but for sure in an AB/B amp. The signal causes a current which causes a voltage sag that corresponds... It's not uncommon to have an older cheap low powered amp that has a change of 20 volts or more from silent output to full blast. The capacitors try and decouple that signal, but there's still traces of "music" on the power supply because of it.

Ahh. I see. Somewhere in my mind I had intuited that the varying audio signal would cause fluctuations in the power supply, because essentially a changing audio signal is a changing load. The "load pattern" of the audio signal is reflected in the power draw. Is that about the gist of it? I know I probably use the wrong words but I can see it in my head this way.

So, if that is the case then I am correct that the audio signal does not flow through the rectifier, but the audio signal itself creates variations in the power supply demands (and hence the rectifier) which are potentially audible.

I think I got it... (?)