SET Asylum

Single Ended Triodes (SETs), the ultimate tube lovers dream.

Return to SET Asylum

Message Sort: Post Order or Asylum Reverse Threaded

Why Single ended ?

71.142.240.49
Posted on January 7, 2010 at 19:51:22
coffee-phil
Audiophile

Posts: 1444
Location: Shingle Springs CA
Joined: January 7, 2010
 Please don't flame me, I'm just asking. I know why triodes are prized over pentodes or beam power tubes and why people like class A amplifiers over class AB but I fail to see the advantage of a single ended power amplifier over a push-pull amp. Push-pull amplifiers can and have been built with triodes and they can and have been biased class A.

This trend seems to be growing so there must be a reason. Please enlighten me.

Phil

 

Hide full thread outline!
    ...
Simple ... better sound!, posted on January 7, 2010 at 20:13:02
Naz
Audiophile

Posts: 2184
Location: Sydney
Joined: September 2, 2005
 At least that's the common opinion of most on this forum. Do a search or better still go listen to a half decent SET. "Immediacy, Holographic, 3D, more like real music" ... these are all terms that are commonly used to describe SET sound.

True, on paper PP looks good and I've bought, built and heard some damn good ones but for me; I just love my SETs!

Naz

 

RE: Why Single ended ?, posted on January 7, 2010 at 20:35:26
casouza
Audiophile

Posts: 755
Joined: March 18, 2008
 For me it was love at first listen. MUSIC instead of Hi-Fi. No specifications can explain it.

Having owned several top-rated SS PP amps (BEL, GAMUT, McIntosh, ASR Emmiter) plus ARC and McIntosh 65-90 WPC tube PP amps, a pair of 845 DHT SET monoblocks made the musicians materialize in front of my listening couch.

That can be quite a heart-warming experience if the singer happens to be Diana Kral, Stacey Kent or Melody Gardot.
Coltrane on SETs is probably a better trip than mind-altering drugs.

Better still, poor recordings are more listenable, because SETs do the "emotion" stuff, they highlight the inherent musical soul of the performance.

Erroll Garner's Concert by the Sea is a good example. It is an awful recording with plenty of musical magic. PP amps tend to emphasize the Hi-FI side, rough sound, muffled highs, where is the drummer?... while SET amps highlight the emotion and relentless swing hidden under the grungy recording.
One can almost hear Erroll Garner sweating...as a matter of fact he had to go backstage and change his clothes twice during the concert.

There are some downsides to SET amps: bass is not so tight; dynamics are not as punchy as high-powered PP amps; heavy-metal, techno and hip hop fans must look elsewhere and one must find compatible speakers (over 90 db/watt, impedance curve close to 8 ohms and well-behaved).
Modern SET amps are not frequency-response challenged...and fortunately there is a trend (and the technology is available) to make more efficient speakers.

I hope this helps explain the SET magic!


 

RE: Why Single ended ?, posted on January 7, 2010 at 20:54:51
Paul Joppa
Industry Professional

Posts: 7295
Location: Seattle, WA
Joined: April 23, 2001
 Naz and casouza have already said it. My own experience agrees with them. Theories abound as to why, but I'm not aware of any really solid scientific evidence, one way or another. That's kind of frustrating, since I'm an engineer by trade and a designer by passion. But that's the reality, for me.

I've heard a few excellent P-P tube amps, and if they are done really well (no feedback, triodes, Class A, good transformers, ...) they can be on a par with SETs for overall quality. But they still don't do emotional communication as well - they make up for it in transparency. If you're an analytical sort, a mastering engineer for example, they may well appeal more or at least be more useful. But I want my system to help communicate the musical emotion from the composer and performer to my head. And feet! SETs still do that better than anything else.

 

RE: Why Single ended ?, posted on January 7, 2010 at 21:02:59
kurt s
Audiophile

Posts: 1137
Location: California
Joined: October 12, 2009
 If you're looking for a possible objective reason, the best one to date has been a comparison of human ear distortion to SE harmonic distortion in terms of its spectrum. It is the topology that most replicates the ear and the ear likes it just find, to a point it can adapt enough to not hear it anymore. SET distortion is on par percentage-wise as the human ear as well. All other amps are most dissonant to music and the ear in comparison.

Music was created originally with acoustic instruments that the human ear perceived harmonious to it. Death Metal was created to ruin the human ear for life, OTOH. So the best amp for Death Metal would be class BC amps of harshest distortion spectrum possible and the most traumatic to the ear and brain possible. So it really depends on the music you like, also.

-Kurt

 

No phase inversion needed, posted on January 7, 2010 at 23:34:50
drlowmu
Manufacturer

Posts: 9730
Location: East of Kansas City
Joined: January 10, 2005
 S.E. keeps the music intact. Simpler circuitry, messes less with the music.

KISS

Jeff Medwin

 

For me it's the speed and microdynamics..., posted on January 8, 2010 at 00:48:34
kuma
Audiophile

Posts: 10272
Location: IN
Joined: July 8, 2001
 Plus all the usual positives that are associated with SET amps.
A simple cct. seems to be more faithful, unadulterated and easier to track more nuances of the music than many others out there.

I haven't heard anything better for reproducing human voices.

I'm no connoisseur, but even I get it.

 

Personal preference for the sound, posted on January 8, 2010 at 02:08:20
Frihed89
Audiophile

Posts: 15703
Location: Copenhagen
Joined: March 21, 2005
 With my speakers (L3/5As) nothing sounds better--more like music-- or more relaxing to my ears.

I also like class A PP amps because the music "feels" more powerful and clean = more real than from A/B PP amps. This would be my second preference.

 

It is different., posted on January 8, 2010 at 03:30:24
PakProtector
Audiophile

Posts: 12361
Joined: May 14, 2002
 hey-hey!!!,
SE amps sound different. Some like that difference, others...not so much. If you go looking for well-done amps, Class A power stages, big power supplies and good parts, etc. I suspect you'll find much more SE built in that fashion. Actually going to the trouble of designing a thoroughly Class A output stage in PP is not exactly trivial, or light...:) Now if you want to go evaluate the amps side-by-side, a class A PP amp is going to be hard to find; everybody knows that SE sounds best so why make the effort on PP...heh-heh-heh
cheers,
Douglas

Friend, I would not hurt thee for the world...but thou art standing where I am about to shoot.

 

Because we can and we want to? (nt), posted on January 8, 2010 at 03:33:00
Thorsten
Manufacturer

Posts: 4209
Location: Somewhere nice on planet dirt
Joined: September 25, 1999
 No Text!
Thor

At 20 bits, you are on the verge of dynamic range covering fly-farts-at-20-feet to intolerable pain. Really, what more could we need?

 

BTW..., posted on January 8, 2010 at 03:36:22
Thorsten
Manufacturer

Posts: 4209
Location: Somewhere nice on planet dirt
Joined: September 25, 1999
 While I generally prefer SE Amp's I now listen to a PP Amp which in many ways was designed following similar principles as I would apply to SE Amp's. The result sounds very good and has more power than most SE Amp's, which my current Speakers require.

Ciao T
Thor

At 20 bits, you are on the verge of dynamic range covering fly-farts-at-20-feet to intolerable pain. Really, what more could we need?

 

Would you be a little more specific please?, posted on January 8, 2010 at 03:52:31
Vatron
Audiophile

Posts: 455
Joined: March 11, 2001
 What ever you would like to share about such a PP design would be appreciated.

NW

 

RE: Would you be a little more specific please?, posted on January 8, 2010 at 04:04:53
Thorsten
Manufacturer

Posts: 4209
Location: Somewhere nice on planet dirt
Joined: September 25, 1999
 Hi,

I am talking about it here:

http://www.audioasylum.com/forums/tubediy/messages/17/178605.html

It is really simple.

Input is ECC83/12AX7 single section in classic resistor loaded Amp (390 Ohm cathode and 330K Anode IIRC) directcoupled to a 6N12 (kind of like an octal 5687) with both sections in parallel as concertina Phasesplitter.

I spend quite some time touching up the operating points so the output stage clips before the front-end, this direct coupled circuit could use a higher +B than I have.

Output stage is what Menno Van Der Veen calls super triode (I call it double local feedback pentode, just for the heck of it). It is basically Audio Research's output stage with ultralinear and cathode feedback from the output transformer secondary and fixed bias. Coupling Cap's are 0.1uF silver mica per output tube. Frontend PSU decoupling uses film caps.

Output Valves right now are Svetlana 6550 at 360V/80mA.

There are no global feedback loops, only the feedback applied through the output transformer to the output tubes.

Produces around 30W at a few % THD, clips very gently and sounds quite good.

Ciao T
Thor

At 20 bits, you are on the verge of dynamic range covering fly-farts-at-20-feet to intolerable pain. Really, what more could we need?

 

I agree with Douglas., posted on January 8, 2010 at 06:07:13
KlEo


 
PP or SE?
Coffee-Phil, do you want "human" drive in the Rolls-Royce or express drive with Ferrari? It depends on your individual predisposition.
Only, do not use 12AX7/ECC83 in amplifier input. This is not good.

 

It is mainly speaker dependent, posted on January 8, 2010 at 07:58:24
Chris O
 If you put a SET on a speaker with wildly swinging impedance and low sensitivity you will think that SET is horrible.

Discussing good or bad, SET versus PP, is like discussing car engines without considering the chassis they are put in. You could have a jewel of a high reving 4 cylinder lightweight engine that makes 180 horsepower with a perfect transmission that performs like crud in a Ford Exposition 6,000 lb supertanker. The same thing is true when you try to shoehorn a heavy blown 750 hp rat motor into a Miata, fun though. (See the link for an extreme example of crazy.)

Even low powered PP amps have beter control over woofers that need more control. It is not the SETs fault the woofers need extra control. On the right speaker, the simplicity of a good SET can't be beat.

 

RE: Why Single ended ?, posted on January 8, 2010 at 08:58:52
JimL
Audiophile

Posts: 3773
Location: New Mexico
Joined: November 24, 2002
 Some of the theory behind why SET amps MAY sound better was discussed in a Master's Thesis by Daniel Cheever. To summarize, the ear itself produces harmonics when exposed to a pure tone, and the higher the volume the more the harmonics. Non-feedback triodes produce a similar harmonic spectrum, resulting in masking of the triode's harmonics by the ear's own production of harmonics, leading to subjectively purer sound despite higher THD measurements. Makes for interesting reading.

The other potential reason is that SET amps have very low distortion at low power, which increases as power goes up. When combined with efficient speakers as they generally are, their actual distortion is very low at usual listening levels, where they may be putting out a few milliwatts. OTOH, due to problems with noise, usually measurements of THD are done at levels of 1/10th watt to many watts where SETs tend to have higher distortion than other topologies - however, that is also louder than most people listen, except for musical peaks, where the masking phenomenon probably kicks in.

However as several others have said, regardless of the theory, the bottom line is still listening enjoyment.

 

RE: It is mainly speaker dependent, posted on January 8, 2010 at 09:02:10
JimL
Audiophile

Posts: 3773
Location: New Mexico
Joined: November 24, 2002
 Yes and no. If you happen to have a speaker where the impedance and frequency response curves are inverses of each other - i.e. an impedance peak at a frequency response dip, a high impedance output SET amp can actually result in a flatter frequency response overall than you would get with a lower output impedance amp. This is why, for example, speakers that have good bass response with SET amps may sound overdamped and bass-thin with other amplifiers. So, yes, it is speaker dependent (but that is true to some extent for any amplifer-speaker combo) but it isn't NECESSARILY true that a complex impedance curve won't sound good with a SET amp.

 

Because it sounds more real., posted on January 8, 2010 at 12:54:25
samtellig
 We hear single-ended. That's what exists in nature. As Nelson Pass has said, push-pull has no particular basis in nature. The push and the pull never match up.

 

RE: Because it sounds more real., posted on January 8, 2010 at 15:30:47
tuneful
Audiophile

Posts: 190
Location: New Mexico
Joined: October 29, 2004
 "push-pull has no particular basis in nature" Hmm....

Not clear to me that hearing is single ended. Isn't the ear drum pushed and pulled by the alternating pressure/rarefaction of the incident sound wave?

Isn't a loudspeaker cone pushed and pulled by the speaker motor?

Over the years I've heard excellent sounding SET's and crappy sounding SET's, as well as excellent and crappy sounding PP amps. It's probably easier to build an good sounding SET, harder for a PP, especially if your major design goal is high power output. But examples exist of both which sound spectacular, even on real world speakers. It's all a matter of good design, materials, and execution to minimize distortions (especially those not easily measured, but as heard through the speaker).

 

RE: Because it sounds more real., posted on January 8, 2010 at 16:25:43
GTCharlie
Audiophile

Posts: 983
Location: Philippines
Joined: December 9, 2004
 matching the push and the pull, even as the tubes deteriorate differently ... whew ... I'd rather solder an SE amp.

 

RE: I agree with Douglas., posted on January 8, 2010 at 16:54:05
Ron Oehlert
Audiophile

Posts: 5109
Joined: December 18, 2003
 Why is that (per the 12ax7)? Please provide reasons beyond *my opinion*.

 

RE: Because it sounds more real., posted on January 8, 2010 at 18:36:15
Donald North
Manufacturer

Posts: 1296
Joined: February 8, 2001
 Air and the eardrum are asymmetrical in response to compressions and rarefactions. This asymmetry is similar to the asymmetrical behavior of single ended amplifier.

Donald North

 

RE: Because it sounds more real., posted on January 8, 2010 at 22:28:42
Tre'
Industry Professional

Posts: 17293
Location: So. Cal.
Joined: February 9, 2002
 "Isn't the ear drum pushed and pulled by the alternating pressure/rarefaction of the incident sound wave?"

I don't think the ear drum is being pushed and pulled at the same time.

A Class A, push pull circuit, pushes and pulls, at the same time, across the primary winding of the output transformer. One tube pushing while the other tube pulls and then, for the next half wave, pulled and pushed.

An electrostatic speaker's diaphragm is pushed from one side and pulled from the other side, at the same time. Then, for the second half of the wave form, pulled from the side that was pushing while being pushed by the side that was pulling.

But I think the ear drum is pushed by the pressure of air (first half of wave form) and then pulled by the rarefraction (second half of wave form), but not pushed and pulled at the same time.

I'm not sure what this means in terms of audio amplifier design.

Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"

 

Ear distortion from ancient fish gill bones, posted on January 8, 2010 at 22:51:33
kurt s
Audiophile

Posts: 1137
Location: California
Joined: October 12, 2009
 Well, that's what the inner ear bones are, ancient fish gill bones evolved.

The distortion in the ear is also asymmetrical by the bone attached to a nonlinear and non-symmetrical load on it: the cochlea's compression and rarefaction of its inner fluid and the overall structural support irregularities.

I bet drums have some 2nd order distortion from them, too. And eardrums.

-Kurt

 

RE: Why Single ended ?, posted on January 9, 2010 at 01:09:56
coffee-phil
Audiophile

Posts: 1444
Location: Shingle Springs CA
Joined: January 7, 2010
 Thanks for all the replies.

I would like to hear some of these amplifiers compared to other amps especially compared to a well executed Class A push pull amp.

What I am getting from these answers is:

1. The second order distortion matches the ear and is therefore pleasing to the ear.

2. The distortion is vanisingly small at the levels at which these amplifiers are used. This may explain why they are OK but not why they are superior since you could make the same claim for a push-pull amp even with pentodes and biased class AB.

3. Elimination of the need for a phase inverter.
Now I think there may be something here however I have to bleive that it would be simpler to design an acceptable phase inverter than deal with the distortion caused by the output transformer due to the relativly high DC flux density in the core. Of corse this can be mitigated by using a huge transformer core.

I picked up from the answers that Class A push-pull triode amplifiers are excedingly rare so comparisons or SET amps with them don't happen.
The SET topology pretty much forces Class A biasing. I hope I don't hear someone say he likes SET amplifiers biased class AB.

I am guessing that comparisons of SET amps to push-pull amps is also a comparison of class A amps to class AB amps.

Please don't get angry with me. I am not dissing your amps, there must be some reason they sound better to many people and I am trying to find a reason explainable with the laws of physics what it is.

Phil

 

Thanks (nt), posted on January 9, 2010 at 08:08:15
Vatron
Audiophile

Posts: 455
Joined: March 11, 2001
 NW

 

This is more to be experienced than can be explained by physics, posted on January 9, 2010 at 08:30:48
kurt s
Audiophile

Posts: 1137
Location: California
Joined: October 12, 2009
 >>1. The second order distortion matches the ear and is therefore pleasing to the ear.<<

2nd order is just one harmonic. The entire spectrum of distortion counts. These amps tend to match the entire spectrum that the human ear hears naturally. PP amps do not match this spectrum as close. They provide different advantages, like better control of speakers. With PP you can get a better beat in the bass, from SE you get a midrange that is unmatched in matching the ears and a soundstage that is deeper all because of "correct" or "correctable to the mind" distortion.

>>2. The distortion is vanisingly small at the levels at which these amplifiers are used. This may explain why they are OK but not why they are superior since you could make the same claim for a push-pull amp even with pentodes and biased class AB.<<

Class AB distortion is not like class A at all. Class AB will introduce a crossover distortion where one tube is cut off and only one conducts. This makes this type of distortion possibly not "monotonic", meaning the amount of distortion does not always go down with reduced power, but will for a portion go up instead, and then down again.

>>3. Elimination of the need for a phase inverter.
Now I think there may be something here however I have to bleive that it would be simpler to design an acceptable phase inverter than deal with the distortion caused by the output transformer due to the relativly high DC flux density in the core. Of corse this can be mitigated by using a huge transformer core.<<

There is another way to do SET amplification, and that is to use small accurate DC only output transformers and use a DC blocking capacitor to that output transformer so no wasted flux causing current is there. It is debatable which type is actually better. I've done both ways, and both have huge potential for detailed low distortion used at the right signal levels.


>>I picked up from the answers that Class A push-pull triode amplifiers are excedingly rare so comparisons or SET amps with them don't happen.
The SET topology pretty much forces Class A biasing. I hope I don't hear someone say he likes SET amplifiers biased class AB.>>

You are correct here. But a Class A PP triode amp can be found in modern production by some small outfits, the specialty kind if you research enough. LA Audio had a 300B push-pull I don't know if they still have, for example.

>>I am guessing that comparisons of SET amps to push-pull amps is also a comparison of class A amps to class AB amps.<<

Only if you don't obtain a class A triode PP amp, that is hard to find. Normally, your answer is yes, with other extenuating circumstances such as a comparison between triodes and pentodes, a comparison between negatively fed back amps and zero feedback, if only comparing the more common examples of these amps.

-Kurt

 

RE: Because it sounds more real., posted on January 9, 2010 at 09:21:31
tuneful
Audiophile

Posts: 190
Location: New Mexico
Joined: October 29, 2004
 I'm not sure how it relates to amplifier design either. Here's another analogy to consider (bit of tongue in cheek here):

For those of us with two functioning ears, if sound from a mono source hits both ears in-phase. Ideally, both ear drums go in at the same time, and both go out at the same time. Sort of like signals in a push-pull amp stage. Since the ear drum motions are the same, we get no spacial information, which could be termed a non-distorted signal. If we listen to a stereo source, there are differences in ear drum motion from side to side; those differences are distortion in the push-pull aural signal and provide the spacial information we enjoy. Again, no real relation to amplifier design, but the implication is we hear in push-pull.

 

RE: Why Single ended ?, posted on January 9, 2010 at 15:35:26
AmanteDiMusica
Audiophile

Posts: 162
Location: US Midwest
Joined: August 21, 2006
 badly, done, anythingg sounds like crap.
Well done, a SET sounds way better than a well done pentode.
That's all.
Grazie
Amante

 

RE: Why Single ended ?, posted on January 10, 2010 at 06:23:39
Neff


 
Although others have explained it well, I'll add my 2 cents from experience of DIY on a clean sheet regarding SET, SE triode connected tetrode & pentode, PP amps plus modifications of a few SET amps & a lot of PP amps. Keep in mind my builds/mods have been of conventual type with either tube stage direct coupling or RC design.

SET- I believe the overall simplicity of a SET amp makes for better sonics. Less parts, less gets in the way to good sonics with SET. The selection of tubes are important including the circuit as in any amp design. Anything less than the better parts & circuit design could result in a PP amp sounding better over SET (of course).

PP- I never had success with mainstream stock or carefully modified PP vintage amps. I can roll tubes, change type of tube, change the circuit even change the output transformers plus PS mods & a simple SE 6BQ5 Magnavox amp with a few mods sounds better. No power from the Magnavox, but does sounds better.

I have built one PP amp that is very close to a good SET amp and in the bass is a little better than SET. I used PP triode connected 6V6 tubes and a simple 6CG7 driver simular to the Eric Kingsbury 'Music Machine'. The PS was a LCLC to audio transformer B+ tap and another LC to the twin tail phase inverter. I used oil caps & poly types in the PS. Only one type of 6V6 sounded good being the early Sylvania 6V6 small grey plate. The later black plate Sylvania 6V6 and a few other 6V6s tossed the sonics 'out the window' IMO. I also got a small extra tweak due to matching the 6V6s from a case of NOS tubes. I had a load of 6CG7s and the Raytheon 6CG7 black plate sounded just right. Any other 6CG7 tube I had on hand 'busted' the amp sonically. I also unloaded the 6V6s primary impedance to twice the RCA recommended operating parameters. This high impedance was important and distortion products just melted away like a clean glass of water.

What I am saying is I think I got lucky on the DIY PP amp that was rediculously fussy with the type & era of tube manufacture. SET amps are not that critical. (For an example, a RCA 45 & Sylvania 45 both sound good if a little different).

A SET 2A3, 45, triode connected tetrodes & pentodes sound better than every PP amp in every area except the one DIY PP example above that was acceptable to me. Additionally, nothing beats the midrange frequency of a SET. Of course, speakers are vital to match up with SET. Finding the right speaker for SET is not near as difficult to find vs a right PP amp design and components that compliment one another.

So, either PP or SET can perform. SET is much easier to get ultra high performance. I never cared for any vintage PP amp I owned either stock or modified. I have not owned higher end PP amps such as VTL, AR, Citation, Mac, etc so be careful with my statement.

Also note worthy is I have an apparent sensitivity to certain types of distortion that is an unpleasent irritation to my ears. I swear some others pay little attention to nuisances some amps produce or perhaps some never heard a natural sounding amp. And I know some others are even more fussy than myself, but I did sell every vintage PP amp I ever owned and new owners were pleased.

 

RE: Why Single ended ?, posted on January 10, 2010 at 08:36:21
PakProtector
Audiophile

Posts: 12361
Joined: May 14, 2002
 ----I have not owned higher end PP amps such as VTL, AR, Citation, Mac, etc so be careful with my statement.


Sorry to but in, but those are not 'high-end'. They're paperweight quality IME.
cheers,
Douglas

Friend, I would not hurt thee for the world...but thou art standing where I am about to shoot.

 

RE: Why Single ended ?, posted on January 10, 2010 at 14:39:55
jj2468
Audiophile

Posts: 52
Location: SF Bay Area
Joined: April 3, 2008
 Amplifier push pull and sound waves push pull are different things. Your hearing is not a push-pull amplifier. Look elsewhere for a good answer.

Push pull, generally speaking, splits the sound wave into two halves. One above the median and one below. One tube (or more) amplifies the half above the median and another tube (or more) amplifies the half wave below the median (or x-axis). The two halves are re-combined to make the whole wave. The purpose is to obtain greater power by splitting the amplification job among more than one tube and one step of amplification.

An SET amplifier uses a single tube to amplify the whole wave form, both above and below the x-axis (also know as 360 degrees of the wave, meaning the entire wave).

Your ability to form a spatial image of music comes from sound waves reaching your ear drums at slightly different moments, since your ears are a few inches apart from one another. It is not push pull distortion.

Jeff

 

RE: This is more to be experienced than can be explained by physics, posted on January 10, 2010 at 17:49:54
coffee-phil
Audiophile

Posts: 1444
Location: Shingle Springs CA
Joined: January 7, 2010
 Hi Kurt,

Thanks for the exelient reply.

Was there a typo in the part where you mentioned the small output transformer which was DC blocked. You said DC transformer. Did you mean AC only transformer? Is this topology what some are calling a parafeed?

Thanks,

Phil

 

RE: Why Single ended ?, posted on January 10, 2010 at 17:56:30
coffee-phil
Audiophile

Posts: 1444
Location: Shingle Springs CA
Joined: January 7, 2010
 Well Douglass,

My main Amp is still a Citation IV. While I would not call it high end as far as tube stuff goes I hardly would regard it a paper weight.
Most people who hear my stereo system think it sounds good.

Phil

 

RE: Why Single ended ?, posted on January 10, 2010 at 18:23:53
coffee-phil
Audiophile

Posts: 1444
Location: Shingle Springs CA
Joined: January 7, 2010
 Hi Jeff,

What you said is pretty much right-on for a class B amplifier, partially true for class AB amplifiers, but misses the mark for class A amplifiers.

In a class A amplifier the output active device (triode, pentode, beam tube, BJT, or FET) conduct for 360 degrees of the signal wave.

In a tradiional push-pull amplifier with transformer coupling at the output the signal and its invert are applied to two pretty much identical amplifiers whose outputs are combined in a transformer. If the amplifiers are biased class A they both amplify the entire wave. Depending on the gain match and transformer ballance even order distortion will be at least partially cancled.

The push-pull configuration permits class AB or class B operation but does not require it.

In addition to the even order distortion cancelation (not such a big deal with class A) the push-pull configuration mitigates the issue of transformer core saturation by the DC bias current of the output devices.

Phil

 

RE: Why Single ended ?, posted on January 10, 2010 at 23:54:30
jj2468
Audiophile

Posts: 52
Location: SF Bay Area
Joined: April 3, 2008
 Phil,
How can a single tube amplify 360 degrees of the wave and still be called "push pull"? Perhaps I am misunderstanding your explanation. Could you say more or perhaps provide a reference?

I get the partial push pull of class A/B (still, wikipedia calls it push pull). I get the topology that more than one tube can amplify less than 360 degrees, but be configured to act together as a single tube (but wikipedia still calls that push pull).

Last, of course, there is parallel single ended triode where more than one triode tube amplify the whole wave form simulaneously, but I don't think that is what you meant.
Jeff


Jeff

 

RE: Why Single ended ?, posted on January 11, 2010 at 01:56:33
coffee-phil
Audiophile

Posts: 1444
Location: Shingle Springs CA
Joined: January 7, 2010
 Hi Jeff,

I wonder if there is a way to link pictures to these posts. I find myself sometimes handicapped with only words at my disposal with no pictures.

Push-pull describes amplifier topologies where there are an even number of output devices and they are arranged such that during the cycle of the signal the current in one half of the devices is increasing while in the other half the current is increasing. There are several topologies which do this. The totem-pole output stage in solid state amps is a form of push-pull as is the similar output of vacuum tube OTL amps. What most of us tube lovers (not the OTL group) think of as push-pull is the transformer coupled output type. In this topology the plate(s) of one of the output tubes (or group of tubes) is connected to the start of the primary winding of the transformer and the plate(s) of the other output (or group of tubes) is connected to the end of the transformer primary. The center of the primary winding is tapped and is connected to signal ground. The tap is the DC supply to the output tubes but it is signal ground by virtue of the filter capacitor connected from it to ground. The signal voltage at the plate of the tube at the start of the primary winding is in phase with the voltage at the start of the secondary or speaker winding while the voltage at the plate of the tube at the end of the primary winding is inverted from the voltage at the start of the secondary winding.

The grid of one of the output tubes is driven by the output of a voltage amplifier stage and the grid of the opposite tube is driven with an identical voltage amplifier. Somewhere in the path to one of the output tubes is a stage with unity inverting gain such that the phase from the input of the amp to the speaker output is the same. If you were to remove one of the output tubes (or group of tubes) you would have a single ended amplifier. With both of the tubes in place each half contributes (hopefully equally) to the output.

Note that the above description said nothing about the class of the amplifier.

Now lets talk of classes of amplifiers.

Class A. In a class A amplifier the bias is set such that plate current flows during the entire cycle of the signal. The quiescent current is one half the peak current. The average current for the entire cycle is the same with no signal as it is at maximum output power.

A single ended or a push-pull amplifier can be biased this way.
These amplifiers are the least efficient but give the least distortion.


Class B. In a class B amplifier the output stage is biased at the threshold of conduction and is driven on with the signal. Plate current flows for 180 degrees of the signal cycle.

These amplifiers can be single ended or push-pull.
They are not useful as audio amplifiers single-ended but can be used single ended for RF amplifiers even in the case of single sideband or AM with low level modulation.
They are not much good for high fidelity audio.
These amplifiers are much more efficient than class A amplifiers.

Class C amplifier. Current flows for only a few degrees of the signal.
These amplifiers are very efficient but they are not useful for audio. They can be used for RF power amps for FM or AM with high level modulation.

Class AB amplifier. This amplifier is between the Class A and Class B amplifier. At very low signals they are in essence Class A amplifiers and they move into class B for large signals. For audio use push-pull is required. Class AB push-pull amplifiers are useful as high fidelity audio amplifiers and are widely used. Vacuum tube amplifiers are well suited for Class AB operation as the larger even order distortion is canceled by the push-pull configuration. The crossover region is more bothersome with BJTs. FETs promise to be similar to vacuum tubes in this regard but I think this will be argued well beyond the rest of my life.

The take-away from all of this is that for audio Class B or AB pretty much compel the push-pull configuration. With Class A push-pull is optional.

Phil

 

RE: This is more to be experienced than can be explained by physics, posted on January 12, 2010 at 16:29:14
kurt s
Audiophile

Posts: 1137
Location: California
Joined: October 12, 2009
 I did mean "AC only transformers" that are used in parafeed. You got it correct.

-Kurt

 

RE: Why Single ended ?, posted on January 12, 2010 at 16:52:38
PakProtector
Audiophile

Posts: 12361
Joined: May 14, 2002
 Citation V maybe? haven't heard of a tube type four until now.

Just to be expand on the paperweight pov, some of the sought after classics sound pretty good. I've had a fair number of them myself, and consistently find well done DIY to be better...so the classics serve paperweight duty. Not that the DIY wouldn't do better; it usually ought-weighs the classics a few to one...:)
cheers,
Douglas

Friend, I would not hurt thee for the world...but thou art standing where I am about to shoot.

Friend, I would not hurt thee for the world...but thou art standing where I am about to shoot.

 

RE: Why Single ended ?, posted on January 12, 2010 at 17:37:04
coffee-phil
Audiophile

Posts: 1444
Location: Shingle Springs CA
Joined: January 7, 2010
 Hi Douglas,

You are quite right. The amp is a Citation V. The Citation IV is the companion pre-amp which I also own. It has some problems now so I am forced to use a solid state Hafler preamp untill I find time to sort it out. Actually the Hafler is not bad.

Phil

 

RE: Because it sounds more real., posted on January 16, 2010 at 06:05:12
PakProtector
Audiophile

Posts: 12361
Joined: May 14, 2002
 hey-Hey!!!,
---- Since the ear drum motions are the same, we get no spacial information, which could be termed a non-distorted signal.

I don't think that's it at all. The ear is recieving two SE signals, and or internal processor is taking phase differences to do the 'spatial thing'.

In every case I've been auditioning a good SE amp there has been somethig that drives me nuts. The better the SE crowd likes the amp, the more of this effect I hear. IMO there is some sort of non-linearity involved, and I can think of a few reasons behind it.

Take the OPT's airgap for instance. That one has an equally beneficial effect to PP signal Iron. There is no getting away with it for SE, except CCS-rigged parafeed, or flux cancelling( usually with filament current on a dedicated coil ). Now I'm not suggesting full idle current capability here, just enough to effect the inductance at small signal levels instead of the extreme variation one sees as the core gets 'moving'.
cheers,
Douglas

Friend, I would not hurt thee for the world...but thou art standing where I am about to shoot.

 

RE: Why Single ended ?, posted on January 16, 2010 at 06:21:19
Neff


 
I never heard a factory manufactured PP amp I could live with compared to DIY. But, we all have different opinions of what is pleasing to the ears.

 

SE, posted on January 16, 2010 at 06:33:13
Neff


 
I agree SE takes careful design for a sonically neutral amp without a slight hazy presentation that equals irratation to my ears. I do not have enough DIY SE experience, but I suspect as power goes up, the larger output transformers create problems. My DIY SE amps I built are typically 3-watts output. I have built one PP amp I like a lot at 12-watts output. Fussy bast*rd to get right though.

 

RE: Why Single ended ?, posted on January 16, 2010 at 10:16:58
PakProtector
Audiophile

Posts: 12361
Joined: May 14, 2002
 -----I never heard a factory manufactured PP amp I could live with compared to DIY.

That sums it up pretty well. No matter what you want, it is far, far, far less expensive to just do it yourself.
cheers,
Douglas

Friend, I would not hurt thee for the world...but thou art standing where I am about to shoot.

 

RE: SE, posted on January 16, 2010 at 10:20:53
PakProtector
Audiophile

Posts: 12361
Joined: May 14, 2002
 -----...but I suspect as power goes up, the larger output transformers create problems.

That has been my experience as well. Seems a matter of size; for PP it is less than ~70W, which corresponds to about a third of that for SE. Too much stuff moving around with the signal, more parasitics, etc. to sound good at lower levels too.

Most of my listening is done at low power but from time to time I wish substantially more...:) For me, an amp has got to do the low level stuff well just as it has to put down some horsepower.
cheers,
Douglas

Friend, I would not hurt thee for the world...but thou art standing where I am about to shoot.

 

RE: SE, posted on January 16, 2010 at 13:15:14
Neff


 
I been wanting about 15 watts SE. I thought a HV 845 may work, but.......

Then, 300B pushed into A2. Looking at cathode direct driven with cathode choke, but have reservations. Interstage transformer may be the ticket, but what is the right interstage quality wise.

I will get there with a decision.

 

RE: SE, posted on January 16, 2010 at 16:05:32
PakProtector
Audiophile

Posts: 12361
Joined: May 14, 2002
 hey-Hey!!!,
I'm getting a PP 211 amp together. Its cathode drive IT is a 3k:600, both split and CT with U-L taps on the 3k side. Seems that PP will make the standing current stuff go away. No way I'd try this one with SE. I plan to use TV sweeps to deliver the ~15 Watts to the 211 cathodes. Should get nearly 90W out whilst still class A...:) and all from about 600V of B+.

The big benefit, as I see it is not having a step change in power tube input Z when grid current starts flowing. The stage ahead of it is pretty consistently looking into a Class A 500 Ohm CT load...same for an 845, but with a bit less than half the required voltage input required for the 211. Not worried about output Z, this one will be running an efficient horn-loaded driver.
cheers,
Douglas

Friend, I would not hurt thee for the world...but thou art standing where I am about to shoot.

 

RE: Because it sounds more real., posted on January 16, 2010 at 16:28:26
PakProtector
Audiophile

Posts: 12361
Joined: May 14, 2002
 That's one of the more interesting reasons to justify liking/selling SE over PP. Completely irrelevant, but interesting anyway.
cheers,
Douglas

Friend, I would not hurt thee for the world...but thou art standing where I am about to shoot.

 

RE: Because it sounds more real., posted on January 16, 2010 at 16:44:34
tube wrangler
Manufacturer

Posts: 2484
Location: USA
Joined: January 29, 2007
 Hi Doug.

You're onto something here for sure.

I hear this also, and that is why I can't stand most S.E. amps.

And the worse they get, the better the cult likes them.

A S.E. does, however, have one advantage (musically) and that is that it runs a single-ended signal as a whole continuum-- it doesn't split it into two halves and then try to re-assemble it. As far as I can tell, that, and the reliability of not having to re-assemble (no phase inverter, no unequal tubes, etc.), are the only real advantages S.E. has.

You can get around the problem that drives you nuts in S.E. by building the amplifier with a linear T.F.-- or Transfer Function. (this is the term MSU electrical and Audio engineers used when they tested my S.E. amp.... I have called the need for getting it right "transfer efficiency" in the past here on the forum).

The engineers believed that T.F. (plotted across the audio bandwidth for an amplifier), should form a linear, straight line-- whether the amp rolls-off or not is not what this is about. What they want to see is a straight line. If the amp rolls off at top and/or bottom, in a good amp, the T.F. line should still be straight, just tilted in the plot.

They consider T.F. as more important, musically, than any other amplifier parameter. The simple D.C.-coupled 2-stage circuit that I presented measured (in this parameter) far better than anything in their experience-- solid state, tube, chip or test instrument circuit..

S.E. amps can't normally do this, but will instead, have curved portions or spiked portions on the T.F. plot. Mine did have a small spike-- at 60 HZ-- which I (and the engineers) expected. P/P amps can also have these effects, but will normally show a better T.F. plot than a S.E. amp.

Even their best test instruments will not show a straight line perfectly. The D.C. coupled single-ender did-- in this case, surpass the test equipment.

In a S.E. amplifier, this cannot be achieved by ordinary layout and wiring techniques, or by using single capacitors for power supply functions, as one can do in push-pull.

In S.E. amp design, the requirements for P.S. capacitors are far more stringent than any single capacitor can provide.... in both frequency response under the circuit conditions, and for the T.F.

One would expect that by using excellent capacitors with wide published bandwidths (such as Mundorf Silver/Gold/Oil, etc.), for power supply functions, that this would not be an issue.

But, in S.E. applications it is, and that is why I have used 7 different brands of capacitors and 21 different capacitors in only the power supplies for two vacuum tubes, D.C. coupled together.

I find your observation well founded. We have allowed a small amount of A.C. on the input tubes signal and we have found that balanced A.C. heating of the DHT output tube used by us in S.E. operation is what allows the amp to flow music smoothly through the gapped output transformer.

By very careful layout engineering of the chassis, and by very careful routing and sizing (also to lengths) of ground systems wiring, we have been able to avoid any form of hum-bucking, but have been able to keep A.C. hum at the speaker at low enough levels.

Hum at the speaker terminals on a good 98 db/watt 8 ohm speaker will average from 1 1/2 M.V. all the way up to 12 M.V. with an unbalanced DHT output tube. The better that tube is built and filament balanced (at the factory), the lower the A.C. at the speaker terminals will be, and the better and cleaner will be the sound of the amplifier..

The small amount of 60 HZ and 120 HZ artifacts that are allowed to be present actually greatly improve both the transparency and smooth, naturally powerful dynamics of the amplifier as long as the output tube allows a balanced filament structure.

Unbalance in the output tube filament structure, even though the hum that it adds should not hurt the signal, it will hurt it if it is unbalanced. This is not because the hum level is higher, it is because filament A.C. balance has been lost.... in a bad tube which allows this.

I suspect (and this is only my opinion) that the gapped output transformer, and really everything else in the circuit-- needs a constant small signal being applied even when no music signal is applied. A.C. heating does this by supplying a small amount of hum.

This effect seems analogous to what a musician does when he is going to strike a cymbal, etc.,-- he lightly "taps" it ("gets it moving"-- good observation-- yours!) BEFORE he is going to strike it.

Then, when he does strike it, it responds quickly and cleanly-- and fully.

You are very observant to note that this requirement is missing in some S.E. amplifiers. When it is missing, sound energy will "stack-up" until a threshold is reached, then it will "let go". This produces FAKE, and out-of-time-sequence dynamics. This is the opposite of true transparency, and it is also a timing delay that imposes itself onto all kinds of music, and will deliver a nauseating "sameness" to all recordings.

You are right to be bothered by this effect-- it is very real.

A S.E. amp that avoids this problem, and is excellent in all other musical parameters is the ultimate amplifier that you can listen to.

It is also a rare Bird!

---Dennis---

 

RE: Who are 'they', posted on January 16, 2010 at 21:26:45
mach1
Audiophile

Posts: 399
Location: Brisvegas
Joined: April 24, 2005
 'They consider T.F. as more important, musically, than any other amplifier parameter'

Dennis, who exactly are the 'they' you are referring to, and what engineering credentials do 'they' possess. How do 'they' define 'transfer function' in engineering terms ? How do 'they' measure and plot TF ? Please explain how an amp with a rolled off top end like yours can have a straight TF?

Simple straightforward questions, simple straightforward answers please.

Thankyou.

 

RE: Because it sounds more real., posted on January 17, 2010 at 07:37:05
PakProtector
Audiophile

Posts: 12361
Joined: May 14, 2002
 ----A S.E. does, however, have one advantage (musically) and that is that it runs a single-ended signal as a whole continuum-- it doesn't split it into two halves and then try to re-assemble it.


You just don't get it. There is no 'splitting' of the signal into two halves. This illustrates your SE amp salesman PoV quite well. Go try this on your customers, you'll have more success.
cheers,
Douglas

Friend, I would not hurt thee for the world...but thou art standing where I am about to shoot.

 

RE: Because it sounds more real., posted on January 17, 2010 at 16:54:44
tube wrangler
Manufacturer

Posts: 2484
Location: USA
Joined: January 29, 2007

I know the theory, as you do. While it is mostly correct, I just don't buy all of it. Experience has shown that there is much more afoot!

Knowledge is a two-edged sword. When it's right, it still doesn't always apply to everything. When it's wrong, the more people know about it, the more mistakes they will make.

I remain free to advance performance and reliability by questioning and modifying theory as I see fit-- in order to address sonic issues that I deem important musically.

You remain free to advance your thorough knowledge of electronic theory and practice, and to believe all of it whether it is best for a given situation or not.

But don't expect 100% compliance to your viewpoint from me. It's not the best that one can do.

---Dennis---

 

RE: Because it sounds more real., posted on January 17, 2010 at 17:22:10
PakProtector
Audiophile

Posts: 12361
Joined: May 14, 2002
 Now why would I care about compliance from you? You're a salesman of a sort that confirms the salesman's reputation...do as you wish, and when I see you've written things which are serving your sales goals, I'll go and point that out.

-----I know the theory, as you do.

You claim that, yet your writings in explanation of said theory are lacking in the extreme, and full of bits like I called attention to in the first place.

-----Knowledge is a two-edged sword. When it's right, it still doesn't always apply to everything.

What an interesting bit of nothing. Knowlege is not two edged, though its applications may be. If it is right, it applies to that area of which it is right, when you expand the circle and claim that some previous answer must fit there too it seems to me one would be asking for trouble.

-----I remain free to advance performance and reliability by questioning and modifying theory as I see fit-- in order to address sonic issues that I deem important musically.

Still selling amps. Since there is no measure of this 'advancement', I'll remain free to question its existence; certainly you understand that given the level of your 'explanations', doubt will remain. When you actually go and provide a means to confirm by duplication your, errr, interesting claims I look forward to reading them.
cheers,
Douglas



Friend, I would not hurt thee for the world...but thou art standing where I am about to shoot.

 

RE:Do 'they' even exist ?, posted on January 18, 2010 at 15:25:44
mach1
Audiophile

Posts: 399
Location: Brisvegas
Joined: April 24, 2005
 ??????????????

 

RE:Obviously not - Caveat Emptor, posted on January 19, 2010 at 17:37:57
mach1
Audiophile

Posts: 399
Location: Brisvegas
Joined: April 24, 2005
 Total unsubstantiated waffle. Known elsewhere as advertising copy.

At least you're consistent...

 

Page processed in 0.053 seconds.

[ Top Top of Page ] [ Contact Us ] [ Support/Wish List ] [ Copyright © 2024, Audio Asylum, all rights reserved ]
[ General ] [ Speakers ] [ Tubes ] [ Vinyl ] [ Digital ] [ Hi-Rez ] [ Video Asylum ] [ Cables ] [ Tweaks/DIY ] [ Music ] [ Films ]