Are rectifier tubes still relevant? Why would you use one, or avoid using one?

[miklos]

Silicone is an insulator material.

Not if it is forward biased

 

[SY]

Even when forward biased. Or reverse biased. Silicone's flexibility, excellent insulating properties, and thermal stability have led to widespread use as wire insulation and potting material.

I will not make the obvious Pamela Anderson jokes.

 

[miklos]

So I see tube amps driven to max can become compressors. Any time I've used compressors the attack and decay times have a major affect (as do threshold and hold time)on the outcome. And setting them wrong can ruin the music. What are these in a tube amp? I doubt you guys know. And the Chances are that even one of these is randomly set correctly are very slim. Again people take a real phenomenon and barely understanding it, then try to use it to convince people that there is an improvement. Bad compression is worse then no compression.

Compressors are Bad, any shape and form, biased left or right.

 

[miklos]

Even when forward biased. Or reverse biased. Silicone's flexibility, excellent insulating properties, and thermal stability have led to widespread use as wire insulation and potting material.

I will not make the obvious Pamela Anderson jokes.

Why not?
Off topic, maybe?

 

[DF96]

Silicone's flexibility, excellent insulating properties . . .

I have told this story before. Many years ago when telephone exchanges were real telephone exchanges and not just computers pretending to be telephone exchanges, the telephone people started finding that their relays and other contacts were becoming unreliable. They remained open-circuit even when mechanically 'closed'. After some investigation they discovered that the cleaners were using the new silicone floor polish (of course, in those days telephone exchanges had to have gleaming floors). This migrated everywhere, including onto relay contacts. The next time the contact opened the tiny spark turned the silicone into silicon dioxide.

 

[SY]

Yes, the lower viscosity versions are fiendishly migratory. In some facilities, the operators can't use normal shampoos because of the simethicone.

 

[JMFahey]

In additive synthesizers the low frequency oscillator can be used to cause a tremelo or vibrato effect by laying a low frequency wave over whatever else is being played.
Is this similar to the perceived effect you are describing?
Notice I said perceived effect, not asking if a synth = tube amp.

It's similr.
And yes, maybe calling a tube amp a synthesizer is too much, but *many* (me included) consider them "effects boxes".
Which most are, by definition: anything that changes sound audibly and is preferred by users instead of flat/linear response *is* an effects box.
Ok, call it "sound processor", some find the effects box label demeaning.
Just read the posts: many "add something tubed in the audio path to add "tubeyness/smoothness/warmth" ...

-as the B+ sags the amplified signal will be at it maximum so as it rises up the B+ falls down in response. This is likely to cause clipping of the top of the waveform which will dramatically increase the second and third harmonic distortion components.

I suppose if you get it right it can sound "good" in a instrument amplifier, but I can think of absolutely no situation where it would improve a HIFI amplifier.

Shoog

Fully agree that clipping distortion has absolutely no place in a Hi Fi amp (ugh!) .
That said, some harmonics appear when playing loud, we are talking a few % , based on non linearity, not clipping, which may sound "interesting" to some in certain circumstances.

The "sag" that many guitarists think is great, can be created by a simple strategically sized RC on the output of the power supply.

Basically yes.
You address an important point: small supply caps make for a short time constant which dynamically tracks the signal envelope; large caps, even if fed through same resistance, be it tube built in or added resistor to SS diodes, lose that intimate tracking.

Does lowering the rail voltage of a tube amp reduce its voltage gain or is it the same as a SS amp where lower rails just cause clipping at a lower output voltage? If there the same, rail droop doesn't compress anything, just lowers the clipping point. (Which increases distortion at high levels.

Pentodes reduce gain and compress *significantly* because dropping *screen* voltage reduces tube transconductance, directly affecting gain.
Effect is maximum in no NFB amps, such as VOX AC30, Fender Champ, etc. ; somewhat reduced in higher NFB ones (most Fender/Marshall) but in any case NFB is small to begin with.
In elaborate tube amps, such as the best Hi Fi ones fom the 60's, they added NFB at multiple points , even dedicated cathode NFB windings in output transformers, in that case amp is way more linear when clean, but clips sharper.
Of course, they were meant not to clip.

the attack and decay times have a major affect (as do threshold and hold time)on the outcome. And setting them wrong can ruin the music. What are these in a tube amp? I doubt you guys know.

You'd be surprised 😉
My job is to make SS guitar amps which in some aspects sound like tube amps, that's one fundamental aspect I had to study a lot in order to closely emulate them.
Must be doing things right, lots of players have *sold* their Marshall or Ampeg tube amps to buy mine, and use them in Stadiums.
Going steady after 47 years (having started with tube amps, of course) , with some 14000 amps delivered.

And the Chances are that even one of these is randomly set correctly are very slim.

Not really; classic tube guitar amps have been built by the millions since the 40's, designers have tweaked them by ear but the "test lab" is live playing and the "lab instruments" are musician's ears.
Since Musicians vote with their wallets, well designed amps survive and thrive (specially Fender), while poor ones disappear.
Pure Darwinism.

I agree with you entirely in principle, but you're forgetting one important thing: back when these amps were new, musicians created new music using these amps.

Good musicians always find ways to use the weaknesses and defects in their equipment to their advantage. Any good guitarist with an amp with built in compression/expansion will tinker with it, altering her playing technique, until sooner or later she finds something that sounds good, and takes advantage of the flaws in the amp. And that's what will end up on the record.

Years later, other young people are still listening to those records, and want to play the same music. And they find that using the same sort of faulty amp allows them to re-create the original sounds.

So what happens is that these sorts of saggy/squeezy amps become well suited to one particular genre, or one style of music. Often, they become much worse choices for some other types of music.

And that's exactly what I notice: blues players, playing long sustained notes, and striving to put expressiveness into those notes, may like saggy valve amps that compress their guitar signal. But you won't find a lot of very technical players using these sorts of amps; shred and prog rock players want an amp that responds to their lightning-fast picking, not an amp that is slow and saggy and takes a couple of hundred milliseconds to recover from the previous note.

There is plenty of that phenomenon going on, no doubt about it. The world of electric guitar is as full of utter nonsense as the world of audiophiles.

In this particular case, though, the evidence is that there is some truth to the claim. But it's only applicable to some players and some genres of music; having a guitar amp with a soggy, squeezy response is not every guitarists cup of tea, it's not a universal improvement, it's just a niche thing that suits some players and some types of music.

-Gnobuddy

Fully agree.

Just as an example, here's a simple song which clearly shows tasty use of *extreme* distortion, over the top intermodulation, used to good effect.
In the guitar you hear lower strings modulating higher ones, and also each other, producing very low frequencies by beating, which is heard as a tremolo effect.
All under control by the player.
As mentioned above, players have grown used to it for decades, and developed their music around it.
The harmonica is also passing through its own small tube amp, again heavily overdriven, to smooth its sound and at the same time give it character.

If both guitar amd harmonica amps were solid state and overdriven to same amount, sound would be unbearable; if both were played clean, fine, but would lose lots of character.
This Lo Fi , Chicago basement Club in the 50's type of sound is still much sought after for a certain type of music.
Can't straight cut and paste the MP3 here but you can hear it in a band approved diffusion page:
Tidido

 

[jon_010101]

For amps that are class A, with nearly constant current draw, the gentle start up to protect other tubes and capacitors seems worthwhile. A tube rectified choke-input power supply is quite appealing from a long-term reliability perspective. Otherwise, I would prefer a solid state voltage doubler with big capacitors for a class AB amp.

My biggest gripe with tubes is that different rectifiers of the same type often have slightly different impedances and voltage drops, leading to different bias points and "sounds" -- So it is best to tune with a particular tube type in mind.

 

[Jay]

silicone

silicon

Silly con! I wasn't aware of that.

 

[Mr_Zenith]

With the RH84 being single ended, and thus class A, rectifier resistance is moot.

Touche - sort of. Don't forget the `25 V drop, though. But your point regarding class A operation is the primary reason for my sloth, even though disposing of the 10 W of power consumed by the heater alone would save me $0.04 per month. 😀

As far as the soft-start benefit is concerned - another argument I'm staying out of - I'm pretty sure my duo of Sovtek EL84's can take the, um, "punishment". 😉 OTOH, if they were RCA single-plate 2A3s...

 

[Gratewhitehuntr]

Some of the content of this thread allowed me today to answer a friend's questions I could not have 3 days ago. "Why is it that our old tube amps sustained SOOO well when turned all the way up?"
Specifically the contributions by Shoog, JMFahey, and Gnobuddy regarding how sound is influenced.

Bob Richards knocked it out of the park concerning implementation and values.

The discussion of SS and why a "shredder" would prefer it is also helpful.

 

[TonyTecson]

I virtually never see vintage high-voltage equipment around here, so no chance of salvaging old high-voltage caps from them.

But a few years ago I discovered that some surplus electronics vendors now sell electrolytic capacitors taken from (used) disposable cameras. These are caps that served as the energy storage in the camera flash circuitry.

Because of their original application, voltage ratings are reasonably high, and ESR is low. Also capacitance is really large - more than ten times larger - compared to the stuff vintage guitar amp designs use. For example, 270 uF, 330 V caps.

I now use these in my valve guitar amps (with solid-state rectifiers, or suitable series resistors to limit peak currents if necessary).

A single cap of this size can do a surprisingly good job of filtering the B+ rail for a preamp, for example. The reactance of a 270uF cap is only about 10 ohms at 60 Hz, so if you have a 10k series resistor in the B+ line followed by one of these caps, you can expect 60 dB (one thousand times) ripple attenuation at 60 Hz, 66 dB at 120 Hz, 70 dB at 180 Hz!

If you had a whole volt of 60Hz ripple before the 10k resistor, it will be reduced to a millivolt across the 270uF cap!

The 330 volts rating may not be enough for every valve application, but there's a work-around there, too: use a lower voltage transformer, and a voltage doubler circuit that only puts half the B+ on each filter cap. Now those 330V caps are good enough for anything up to 660 volt power supplies - terrifyingly high, higher than any sane amateur should ever try to work with.

-Gnobuddy

manila, being in a third world country is a dumping ground for discarded electronics, as such, these are stripped down and picked of precious metals for recycling....

still some enterprising of us reuse the heatsinks, semicons, wiring harness and yes caps for repurposing making amps out of them.....

a mana from heaven for us diy'ers.....😉

 

[Gratewhitehuntr]

Now that scrap prices have come down I expect there will be more surplus components everywhere.

 

[TonyTecson]

first world countries keeps dumping them onto third world countries, so that means we get the goodies....

 

[Wavebourn]

i have to see it in person to confirm...😀

It is hard visually, you can't tell is it hard (*n), or soft (*ne) 😀

Back to the topic, I used vacuum rectifiers once, when experimented with HV for B+, but did not have high frequency high voltage SS diodes.

When people say that they sound better in hi-fi amps, it is usually due to wrong wiring. Higher dynamic resistance causes lower charging currents, and charging time periods are longer.

 

[thewwwall]

Hey - coming in from a technician's POV - i just do what the client or spec'd parts and OPs are and all. Anyway, i have to agree that SS rectification is way more compact, efficient and easier to incorporate/regulate on a given plot of real estate if given a blank slate. For purposes of any guitar amps, my main focus, but yep, SAG is sort of easy to mimic, but if it's for instruments (or old/semi-old school vintage circuits i'm now building) everybody wants 5AR4/GZ34s so, that's what i give them!? Sovteks at that although i use the semi-SS tweak for myself, in garage setup, see Eli's trick for that otherwise, SS rects for me in hifi in house, since it gets hot here and I hate throwing heat out only to get air conditioned for sound PS ;^) - Best Regards

 

[Wavebourn]

Use 6Д14П damper diodes, they are cheap and look tubey. 😀

 

[kevinkr]

OT posts removed. All this because of a spelling error. Let's focus on the subject and not crap up the thread. Thank you

 

[cbdb]

Compressors are Bad, any shape and form, biased left or right.

You obviously haven't done much recording.

 

[Miles Prower]

From a design point of view are there REAL advantages to rectifier tubes vs more modern tech?

If you're considering music reproduction amps, the main advantage of hollow state power diodes is the glowey bottle coolness factor. Otherwise, there are none. Instrument amps are a whole 'nother story. The inherently poorer voltage regulation can be useful in that there is automatic compression when overdriven. Another possible advantage could be that the HV DC comes up slowly, giving enough time for cathodes to warm up.

One project I did used the 5U4GB as the main power diode. In that case, I had the PTX ( 650V_CT @ 150mA ); the current rating was just what I needed, but it overvolted badly when operating into Si diodes. Since this PTX also had a 5.0V @ 3.0A secondary, that suggested 5U4GB. Running some figures against the plate characteristics showed that the forward drop was just what I needed to lose some 100V. Better than resistive dropping or a MOSFET dropper. So, yeah, the vintage PTX was designed for a 350V_DC rail with this type.

Otherwise, I use solid state diodes. As for performance, it makes no difference as I have clean DC either way and that's all I care about. As for hearing power supplies, I leave that to the audiophoolz to debate among themselves.

Hi,

A disadvantage of tube rectifiers (except for guitarists) is they
severely limit the value of the first power supply capacitor.

So for hifi it makes the power supply more complicated,
but for guitarists the rail droop is all part and parcel of
the "soul" of the valve amplifier.

This may or may not be the case. Your Si diodes won't complain if the I_surge is several amps, but your vintage PTX just might since these were never designed with surge currents much over 1.0A in mind. Hollow state power diodes are still high voltage/low current devices, unlike solid state. You just might have to limit the size of the reservoir capacitor anyway and include the usual LC ripple filter.

It's not such a problem these days since you can get high voltage electrolytics in larger capacitances than you could back in "the day", so you can still have effective ripple suppression on the DC rail. Hide the "big one" behind a ripple choke and there's no problem. With the 5U4GB, the reservoir was 34uF (two 68uF/350V capacitors in series with voltage balancing resistors) -> 8.0H/150mA ripple choke -> 220uF/450V output capacitor. No prob, and any ripple well below the noise floor while o'scoping the output.