• 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.

Is single ended Class A better than push-pull?


Disabled Account
2005-08-25 5:29 pm
is single end class a better than push - pull?

is single end class a better than push - pull?

(spam site edited out by moderators)


  • berning_gainblock.gif
    2.1 KB · Views: 1,019
is single end class a better than push - pull?

That's a big: "It depends".

Push-pull has certain advantages. It eliminates (or nearly so) the DC magnetizing current, and so makes the design and construction of output xfmrs easier. Balanced operation also eliminates (or nearly so) even order harmonics, making for cleaner output.

Unforch, it also removes the second harmonic, which doesn't seem to do much damage to listenability, and may even improve it. The odd-order harmonic content coming from a P-P stage has higher order harmonics (3, 5, 7, 9, etc.) even if the magnitude is lower, these have a greater affect on listenability. Can you control that with feedback? Sure, NBD.

Does it make a difference? Who knows? It all comes down to listenability, and that's strictly a personal preference. Some like P-P cleaned up with feedback; others prefer raw SE, and no feedback. You can fight about that one forever (and they're doing that over in the SS forum. :D )

mr mojo

2005-06-06 3:58 pm
I agree wholeheartedly about the coke/pepsi thing. In fact, I think a person's previous experiences, state of mind, sobriety level as well as speaker and room interactions will make far more difference than SE or PP.

To add to Miles' post: The center-tapped output xfrmrs of a PP amp, in addition to limiting/cancelling DC current and core saturation (allowing them to be simpler, smaller and cheaper)also cancels out some hum. Feedback in a PP amp will help decrease hum levels even further-with the combined action of the two allowing for less(cheaper) power-supply filtering such as C-R-C rather than using chokes-although there's certainly nothing wrong with using a choke in a PP power supply!

Without the combined action of feedback and center-tapped xfrmrs, hum can be a problem in a SET amp-but can be cured with care.

mr mojo


2003-01-18 7:57 am
Near London. UK
No, no tolerance whatsoever...

Coke/Pepsi? We are talking about totally artificial hyped-up junk drinks made of water and artificial additives sold at an inflated price and consumed mostly by kids aren't we? Wouldn't a comparison between champagne and cava be more appropriate, or sauternes and graves? Something real... Please?
I am always intrigued by this argument (was this opened mischievously?). In addition to factors mentioned by Miles and Mr Mojo:

Okhifi did not mention triodes specifically. A pentode will always have substantially more distortion (and very audible) than any p.p. circuit, going into high order harmonics. Then the removal of even order harmonics seems to be a stumbling block for some. One must see that correctly: It is not even harmonics per se that are removed, only even harmonic distortion, i.e. something added by the amplifier that should not have been there in the first place. (There is sometimes a perception that part of the original music is "removed".)

If looked at correctly, I for one cannot see that any circuit causing lower distortion, whatever the content, can be inferior to one causing higher distortion, when the original purpose is to have no distortion - that is not logical. (This with the proviso that one is aware of the listener fatigue caused by severe high order harmonics - I am excluding that for the moment.) Thus if a single ended triode is preferred because it has mainly 2nd harmonic distortion, then use a p.p. triode output stage; it will have even lower distortion, etc. Pentodes generally can be capable of generating some higher order products (they will do more so in single ended) but that depends on the design. However since the days of ultra-linear (rendering largely triode type distortion) I cannot see anyone using straight pentodes any more.

That is it, in a nutshell, for me. . .
I read this earlier and ignored it, but now I decided to add my 2 cents. Yes this is largely a matter of personal preference, but it is also greatly influenced by the type of music that you prefer. It is also influenced by the type of speakers that you own. I have several tube amplifiers. I currently have three set up so that I can easilly change between them. I use a SE with 45's (or 300B's), a SE with 845's (or 211's) and a Push Pull with 300B's (no feedback). The 45 excels at the subtle details in most music, but with my 87db speakers, it won't get loud. The 845 SE gives up some detail in exchange for lots more power (40WPC). The 300B push pull (30WPC) has punch and dynamics, even though it has less power, that the SE amps don't have. This amp positively excels for Dance, Techno and some rock music.
My Yamaha NS-10 speakers like the SE amps, but my old AR-4x's like the PP amps. The AR's also like PP pentode amps with feedback.

I have built at least a dozen different types of amplifiers in search of that magic amp that combines the best of all of these without the shortcomings. I have not found it yet. The 833 SE may do it though.

Just because I like SE amps, doesn't mean you will. Everyone has different ears and preferences. Listen and make up your own mind.


2004-01-27 8:55 am
However since the days of ultra-linear (rendering largely triode type distortion) I cannot see anyone using straight pentodes any more.

I tend to agree with you and that's what I would try if I had suitable OPTs. The other form of distributed load, cathode coupling, might be even better. However, my OPTs don't have UL taps or an extra winding, so I'm stuck with either triode or pentode mode (currently using triode mode). Maybe one day I'll get to Singapore and treat myself to a pair of good quality trannies, but I don't have the spare cash yet.

A problem with straight pentodes, quite apart from their inclination to produce odd harmonic distortion, is their very high internal plate impedance (15k ohms for EL34, compared with 910 ohms triode-strapped; UL would be somewhere in between, maybe around 2k?), requiring at least 26dB of NFB to get a damping factor comparable with triodes using no NFB and 30dB or more would be better. That's not to say it can't be done but you need a very high open loop gain to start with and the risk of instability is high. With UL, 10-20dB of NFB would be ample.


diyAudio Member
2004-06-07 8:25 pm
Brooklyn Park, MN
HD, seriously, if you see something that looks like spam, hit the "report" button. Moderators can't be everywhere at once, so member reports are really, really helpful to us.

Sorry SY. I didn't actually click on the link in the first post earlier today because I just assumed the guy was an idiot trying to stir things up (in a very unimaginative way). I think the same guy started another account and has a thread here:


SY, i would be interested in your preference other than SET. what are the SET characteristics that you don't like, or the PP characteristics that you like, and so on. If these have been posted before, I'd be interested to know the URL and read about them.

And no I'm not a troll but just a SET guy wanting to understand the preferences of a PP (or NON-SET) guy.

I'm definitely not closing my ears :D on a specific topology so I'm interested to know what I might be missing.

If you think it is more appropriate to be a private mail, you can send me one, and if you feel you don't want to talk about it, that's also fine with me. I respect whatever decision you come up with.

Don't worry, I've been quite upfront in the past about my reasons for disliking SETs, so I'm unlikely to set anyone off that I haven't already set off. Distortion and source impedance lead the list of my objections, followed closely by ridiculous sensitivity to power supply, output load, and sometimes even the source impedance to the input. My response to EC was figurative, but does reflect my impressions; I've always compared using a tube amp to driving a '57 T-Bird, which would be a blast on the freeway, but a Model T doesn't belong out there.

The inefficiency is just irresponsible. Might as well get the biggest damn SUV money will buy.

One more thing: I strongly believe that if you're going to do something irrational like use a tube amp, it ought to at least be well-engineered.