Rectifier Myths

[Deang]

From the KR website:

Vacuum tube rectifiers have three major parameters and some secondary considerations. The main parameters are peak inverse voltage rating, maximum rectified current and internal resistance. Generally speaking the rectifier for any particular amplifier will be selected on the basis of the first two, i.e. what voltage and current are required.

The myths are that some rectifiers produce a better sound than others.

How can this be?

In a properly designed Class A amplifier the output valves draw a constant current from the HT supply regardless of the output power, likewise the low power stages should have their HT supplies adequately decoupled and filtered from the main HT. Therefore the load on the HT supply should be constant and stability of HT voltage against load current should not need to be considered.

Provided that there is adequate filtering between the rectifier and the amplifier to remove all the supply frequency components from the HT and sufficient audio frequency decoupling so that the HT supply presents a very low impedance over the audio frequency range the amplifier itself should not be able to see the rectifier. Whatever type is used therefore should make no difference whatsoever to the sound of the amplifier.

There are other parameters to be considered when choosing a rectifier such as the maximum peak repetitive cathode current, which will determine the type of supply frequency filter and the value of the reservoir capacitor. The resistance of the high voltage windings of the supply transformer will also need to be taken into consideration. Whether the tube is directly or indirectly heated will affect the time taken for the HT supply to reach its operating value.

It should be noted that the above comments apply strictly to amplifiers with pure Class A output stages. For Class AB or Class B amplifiers in which the load current varies with the level of sound output the design and considerations for the high voltage supply are far more complex and under those circumstances the choice of rectifier could have some bearing on the sound quality of the amplifier.

[mobile homeless]

Lordy. Great reference. KR? Why? KR???? Mysteries will never cease.

The stories about KR would make a small novella.

kh

[Allan Songer]

What a load of crap! I have many different 5U4 variants around here for my MC-30s-- Sylvania 5U4G (1940's big, shouldered envelope), Sylvania 5U4G (1950's, square shouldered), RCA (1950's, tall bottle), RCA (1960's, short bottle), Tung-Sol 5u4G (1940's, Shouldered), Raytheon (1940's, big bottle, shouldered), GEC U52, Brimar 5u4G, WE 274B, Sylvanina 274B. I think thats all of them.

If they all sounded the same why would I have kept experimenting over the years? Because I'm delusional?

That's what Kron would say I suppose.

Anyway, the BEST sounding ones are the Brimar 5U4, the early Sylvaina, both of the 274B and the GEC U52. Hate BOTH of the RCA, the Tung-Sol is supposed to be the ****, but I think is sounds thin, as does the Raytheon.

The most LUSH is the early Sylvania and the most balanced is the Brimar, but lately I've been using the Sylvania 274B, as it is a little "tighter," exhibits a bit more control and thats what I've been enjoying.

[Allan Songer]

Never mind! I didn't get to the end of the KR essay. I guess the MAC with it's odd-ball quasi-class B topology doesn't apply.

Good night all.

[mobile homeless]

Neither do my Class A Moondogs or a host of other amplifiers I have heard. I have not only heard the differences between different rectifier options, but also the SAME tube from different makes. I am not sure if I agree with the above per say as I STILL think it's in the audio path and subject to review, Class A or not. Differences here still seem to be apparent in high resolution systems whether the PS is adequately filtered, decoupled, or not. This "see science, how can this be" approach is not fullproof by any means as the science and measurements dont always plug the loopholes even though it "sounds" like it should (har).

kh

[NOSValves]

The important part that is being missed here is !

In a properly designed Class A amplifier the output valves draw a constant current from the HT supply regardless of the output power, likewise the low power stages should have their HT supplies adequately decoupled and filtered from the main HT. Therefore the load on the HT supply should be constant and stability of HT voltage against load current should not need to be considered.

I bet I could fairly easily limit the effects of Rectifier tube change in either of your amps by simply changing and adding some filtering in your PS. You may not like the results but just the same it will limit the effects of changing that tube.

Okay blast away !

[Deang]

Why KR? Why not KR? Am I supposed to accept the idea somehow that Dr. Kron or his "Kronies" are not experts in the field of tube, or tube amp design?

I just thought it was a short, interesting read, and I also enjoy sharing my misery. It's just amazing to me how little agreement there is in this hobby.

So, we have this "disclaimer", and what do we do with it?

"Provided that there is adequate filtering between the rectifier and the amplifier to remove all the supply frequency components from the HT, and sufficient audio frequency decoupling so that the HT supply presents a very low impedance over the audio frequency range the amplifier itself should not be able to see the rectifier. Whatever type is used therefore should make no difference whatsoever to the sound of the amplifier."

Simply put: Does this mean that if there are audible differences between rectifiers in a Class A circuit, that the perceived differences should be put down to a design flaw?

[Gregorius]

Good Ol' Dean stirring the pot again.

[NOSValves]

Simply put: Does this mean that if there are audible differences between rectifiers in a Class A circuit, that the perceived differences should be put down to a design flaw?

To some extent yes !! Especially in vintage gear although I would not call it design flaw as much as available parts of the era and space to do the job right. These amps were all built to a price point is also a valid. With amps like Kelly's I'm guessing that I could limit the effects not really sure there !

Craig

[NOSValves]

Kelly,

What is the value of the filter caps before the output transformer is fed on your moondogs ? also is there a choke and if so what is its value ? Just curious !

Craig

[mobile homeless]

It's STILL IN THE CIRCUIT, filtered or not. And too much filtering can be just as much a problem as too little (not to mention the filtering method/type/etc). There are a ton of theories on this alone. So far, I have heard a difference in rectifier choice in just about every amplifier I have had the pleasure of testing.

Why not KR? See their amplifiers and tubes... heh.

I suppose KR has problem solved for his products (ie, I love SS rectification). Everyone I have talked to that has heard KR amps in person groans; then again, I have only talked to a handful.

kh

ps- Well, you can go to 100 different sites and get 100 different viewpoints on this subject. Ditto on the forums. Manufacturers have the most to gain, of course.

[dodger]

Greetings:

I agree with Dean.

With all the tube amps at almost any price, if you use a rectifier tube, you will have the "sound" of that PARTICULAR tube.

I have used RCAs with excellent success, Genalex U-52s that are killers, but few have heard of them, and Sylvanias that have sounded poor in home use but graet in guitar amplifiers. Not all Sylvanias, but some.

Even SS is subject to the materials used and the tolerances in specifications.

In my reading of Dean's post, I paraphrase it to "There's theory and then there's reality."

Win dodger

[Deang]

"... suppose KR has problem solved for his products (ie, I love SS rectification). Everyone I have talked to that has heard KR amps in person groans; then again, I have only talked to a handful."

This of course means I would probably think they sound great.

"...Well, you can go to 100 different sites and get 100 different viewpoints on this subject. Ditto on the forums..."

And not just this subject. To be truthful, it really eats at me some. I can understand there being disagreement regarding the sound of things, but one would think from an engineering standpoint -- there would be a wrong way and a right way.

Something I meant to post about several months ago and simply forgot about until last night is Lynn Olsen's preference for high quality SS preamps in front of tube amps (I guess I should limit statement to the just say the Amity). Interestingly enough, he doesn't use an argument built around the idea of better sound -- but one built around simple engineering ideas.

[NOSValves]

So far, I have heard a difference in rectifier choice in just about every amplifier I have had the pleasure of testing.

I would venture to say that almost every amp could easily benefit to MY ears from better PS filtering. I have almost across the board positive results from beefing up the PS in vintage gear the dynamics just come to life and I'm not just talking Bass here its across the board improvement in dynamics and control. Can you install to much sure ! but chances are in most cases there is to little ! Dean's QuickSilvers were a classic example when he ask the designer what could be upgraded what was the answer ! higher value B+ caps !

SET amp lovers in my opinion would not like the result of a beefed up PS becasue of there listening preferences by doing so you will lose some of the sound that you strive after. But some of us are not after that type of sound.

Craig

[mobile homeless]

Jeeez, dean. There are TONS of simple engineering ideas and TESTS from simple engineers that say all sorts of things. And specs and measurements can be used to back all sorts of findings. How do you equate the "solid engineering" behind many of the 70s SS amps with excellent specs they thought meant excellent, perfect sound? Why are you still hung up in this? Use your ears. Keep reading. When you see the words "based on solid engineering or science" as a reason dont assume this is true.

There are so many variables and testing/measurements and even science is still tweened by/for humans with the results not always matching the conclusions drawn forth. It is quite easy to perform a test. It's a hell of a lot harder to USE the results or even fully comprehend where the tests arrive at the listening door.

See Lynn Olsen's speaker designs. See the Klipsch speaker designs. He appears to love examples of both although surely not a fan of Klipsch in the past by any means (based on his posts). I suppose one is wrong? The variables are amazing.

kh

ps- For those of interest, an older Moondogs parts list (it's been changed several times)

A Moondog Parts List example

[dodger]

Greetings:

Back to Dean again.

If you can remember a company called Sybron, nothing to do with Audio, I had the Head of their Human Resources Department as an Instructor for one of my Personnel Classes.

There are many Personnel textbooks that tell you the "correct" way to handle situations. Theories.

However Dave Kausch would warn us, "this is what the textbook says to do.... In the real world, that could not be done...."

He would speak of employment, discrimination, privacy, human temper, ego and a host of other things to show that the answer found in perhaps twenty (20) textbooks (theory) could not work due to one or more of the aformentioned items (reality.)

One can get 100 different answers because we are human and materials that are the same can perform differently under different circumstances.

Tolerances, material quality, hearing curve, room design, room furnishings, room dimensions and a host of other things, even humidity, can and do affect the way something sounds or is perceived to sound. Pre-conceptions also enter the factor.

Theory and equations are the beginning point. The actual sound and sonic performance are the ending except for discussions and viewpoints.

Win dodger

[dodger]

----------------

On 9/8/2003 8:57:53 AM mobile homeless wrote:

Jeeez, dean. There are TONS of simple engineering ideas and TESTS from simple engineers that say all sorts of things. And specs and measurements can be used to back all sorts of findings. How do you equate the "solid engineering" behind many of the 70s SS amps with excellent specs they thought meant excellent, perfect sound? Why are you still hung up in this? Use your ears. Keep reading. When you see the words "based on solid engineering or science" as a reason dont assume this is true.

There are so many variables and testing/measurements and even science is still tweened by/for humans with the results not always matching the conclusions drawn forth. It is quite easy to perform a test. It's a hell of a lot harder to USE the results or even fully comprehend where the tests arrive at the listening door.

See Lynn Olsen's speaker designs. See the Klipsch speaker designs. He appears to love examples of both although surely not a fan of Klipsch in the past by any means (based on his posts). I suppose one is wrong? The variables are amazing.

kh

ps- For those of interest, an older Moondogs parts list (it's been changed several times)

A Moondog Parts List example

----------------

Greetings:

There is a part of all of us that wants and or needs an absolute.

Even in this thread Mr. Songer starts with noting a "lot of crap," then changing that to be ignored with the way McIntosh circuitry performs.

A forceful theory, then a reply to ignore it. An example of theory followed by reality. Though an excellent list of Rectifier tubes was put forth and they do have an effect. They do change the sound.

I cannot and will not attempt to state what Mr. Songer means by "lush." Even in our vocabulary, we have different meanings. Mr. Songer gave an opinion based on his tests.

I would not presume to call his experience a load of crap. He can back it up.

I own McIntosh, I too have heard a difference, as did Mr. Songer. That's two (2) out of just this thread. Reality vs theory.

Theory gives us the imagination to push the limits of reality.

Win dodger

[mobile homeless]

Win, I didnt know this was Dean's major point in the QUOTE from the KR website. HE is saying that with a properly filtered/decoupled PS, you SHOULD NOT BE ABLE TO HEAR the difference in rectifiers.

Some overly regulated PS tend to bring a sound not to some's liking much as Craig says. Also, when using tube rectifiers, it seems it becomes even more problematic. I guess the filtering DOES tend to mitigate the rectifier's affect, to what degree variable. On the other hand, there are other factors to consider.

Jeff and I were exchanging some mail on this. He brings up the good point,

"What's the best, a slow well filtered PSU where all rectifier sounds equal OR a fast nimble "adequately" filtered PSU where the rectifier will have more of an influence.

I'll vote personally for the 2nd options for the "simple SET designs."

Again, we are back to the many, many variables going on. It's what makes it so interesting and so frustrating for those that want it all simple and anwered, so you can fold it up in your wallet.

kh

ps- Win, also see the idea of Sliding bias Class AB/B reference. Allan mentions this in his second post in relation to the KR article Dean Quotes.

[NOSValves]

From the parts list I would say the Moondogs have a pretty stiff PS for 3.5 watts. The differences in rectifiers must really sift through with this design somehow. How much difference using the same type of rectifier are we talking about with these amps ?

Craig

[Painful Reality]

----------------

On 9/8/2003 9:38:37 AM mobile homeless wrote:

Jeff and I were exchanging some mail on this. He brings up the good point,

"

What's the best, a slow well filtered PSU where all rectifier sounds equal

OR a

fast nimble "adequately" filtered PSU where the rectifier will have more of an influence.

I'll vote personally for the 2nd options for the "simple SET designs."

----------------

Yep,

And I'll stand by this considering the case of "simple SET design" in the genre that Wavelength, Welborne, Wright, etc. build

Too much filtering will sound just "dead" to me. But it must be said that the wimpy PSU of a simple SET may appear as a joke for many. I personally rarely use more than 100 uF total capacitance in my PSU (though I rarely use less too).

I am not absolutely commited to tube rectification though and my still favorite amp, the Tenor dual furnace use SS rectification (well implemented it must be said).