Tube Amps: Single-Ended vs. Push-Pull and Buying Advice

I describe the two main types of tube amplifiers and the pros and cons of each. I also give some pointers to potential consumers.

[@davegleeful]

Really enjoyed the video. Don`t think I`ve ever had tube amps explained in such a straight forward and easy to understand way. Ordered a small single ended tube amp for headphone listening a couple of weeks ago and looking forward to listening and tweaking. Thanks for all your other videos, always interesting and entertaining.

[@markfischer3626]

Many people like the single ended vacuum tube triode amplifier without negative feedback. For whatever reason it is very popular with some audiophiles. This is the most primitive design that exists. "Lee De Forest invented the audion, a vacuum tube device that could take a weak electrical signal and amplify it into a larger one. The audion helped AT&T set up coast-to-coast phone service, and it was also used in everything from radios to televisions to the first computers." The 300B vacuum tube which is also very popular was I think invented by Western Electric division of AT&T. The triode was an improvement to the vacuum tube diode which consisted of two plates, a cathode that emitted electrons when heated and a plate that attracted electrons. Forming a complete circuit it only allowed current to pass in one direction. Placing a third element, a control grid that is like a screen between them allows a small change in voltage between the cathode and screen to effect a large change in the current passing between the cathode and plate. This is how all vacuum tube amplifiers work. The amplifier is invariably low power because there is always current passing between the cathode and plate regardless of the control grid voltage. This is called class A operation. There are problems with this design besides its low power output. A small change in the temperature of the cathode results in a much larger change in its electron emissions. The characteristics of the amplifier are therefore highly dependent on temperature. This is why some users will allow their amplifiers to be on for an hour before it is "ripe" for listening to. (negative feedback can correct this problem.) Another problem is that the source impedance of the output circuit is very high so an impedance matching transformer is needed to lower the impedance to match the low impedance of the speaker. It also blocks the DC bias necessary to operate the tube from damaging the speaker. I've tried to be objective here but frankly it's not my choice and personally I don't like vacuum tube amplifiers at all. About 15 years ago I went to a trade show sponsored by Vacuum Tube Valley in Piscataway NJ. To my amazement, by sound alone I picked out the only solid state amplifier of all of the amplifiers at the show. To each his own.

[@paulfalke6227]

Hello markfischer3626, you like solid state amps. Why not? A (good) tube amp has a "sound". Now you can like this sound or not. A good solid state amp sounds "neutral". Some people like neutral. You are not the only one with this taste. Be happy.

[@markfischer3626]

@paulfalke6227 Solid state amplifiers do not necessarily sound alike. The reason is that in a circuit the amplifier will have a different frequency response than it does on a test bench with a resistor for a load and only measured at one watt. The overall system transfer function, loosely comparable to the frequency response from the recording microphone to your ears is what determines in part tonality. Clealy one size doesn't fit all but in modern so called high end systems usually that's all you get.

[@user-od9iz9cv1w]

Well done. Accurate, simple to understand and common sense recommendations. Perhaps the best SE vs push pull explanation I have heard. Also touching on the topic of quality of parts and transformers as a key ingredient. Also love the switch and jack locations. I built my push pull triode monos using all these same concepts and they sound great. Form follows function and the power is at the back. Signal comes in at the front on one side and out at the front on the other side. So I built them as mirror images and set the source in the middle. Signal starts in the middle and gets amplified to the left and the right with the amp near each speaker. Looks great and is functional.

[@jukingeo]

Great video! I know a good bit about tube amplifiers myself and everything you said is spot on. Initially I mostly dealt with push pull amps coming from servicing guitar amps, but for hi-fi, I quickly found out that push pull tube amps really didn't sound much better than a push pull solid state amp. The magic with "that tube sound" extended from the fact that single ended amps have a strong second harmonic, which makes the amp more musical sounding. Also, single ended amps have little or no feedback, which also improves the sound. I raised an eyebrow when you mentioned Skunkie Designs as I religiously follow her page and designs. In fact, I want to build the new 12ax7 / el34 amp she just covered, and is also producing for sale. Her EAR phono preamp and the "color" preamp are also of interest to me. When I found you it was the video you did comparing a single ended el84 amp with push pull 6v6 amp, and even on my computer speakers, I could hear the tremendous sound quality difference the single ended amp had over the push pull one. I know the latter well as I also owned push pull 6v6 amp, but I could never get a hum problem solved with it. Even so, I noticed that is sounded only slightly better than a transistor amp. So I sold it and started a very long research path into single ended amps. Now, I am finally going to build one! A bit off topic, but I am curious to your experience with the Hood solid state amp. Thus is another amp type I am researching, as it supposed to be a warm sounding transistor amp. I would like to team up such an amp with skunkie's color preamp for the ultimate in a warm sounding hybrid integrated amp. Great video and keep up the great work!

[@Lancaster_Hi-Fi]

Thanks! I don't know about the Hood, but now I guess it's better Google it!

[@tothemax324]

John Linsley Hood design is revered by the likes of Nelson Pass etc, diyaudio website have lots of JLH 1969 amp info can be found there. I've owned 3 clones and kept 1. The design is all about the short signal path being a simple/best form of amplification.

[@jukingeo]

@@tothemax324 I was considering Nelson Pass' ACA (Amp Camp Amp) amplifier for a while, and that is a 10 watt (per channel) class A amplifier. A tube pre-amp stage into something like that might be interesting.

[@tothemax324]

@@jukingeo I hear you, I have owned the Freya but being all PCB it didn't quite live up to the hype with me.

[@jukingeo]

​@@tothemax324 There are a couple ways to go with tube sound and I think Schiit opted for a more clean sound. To get "that tube sound" you are going to need some distortion. While it may not look right on paper, sound wise, a strong second harmonic is what gives "that tube sound". That being said, this would rule out push pull amplifiers and amplifiers that use heavy feedback, which is designed to cancel out distortion. Single Ended amplifiers with no feedback are the strongest contenders for "that tube sound". This is why triode tubes, such as the 45, 2A3, 300b, etc are very popular.

[@davegleeful]

Thanks

[@Lancaster_Hi-Fi]

You're welcome! And thank you!

[@petrolfever]

This has been the best video I've watched explaining this topic. Thank you so much :)

[@sonicyouth29]

Thanks man. Best video on the topic on YT.

[@jimamsden]

I understand that a single-ended power amp will generate even harmonics. But it also generates odd harmonics, both due to the asymmetric clipping caused by saturation and cutoff in the tube. In a guitar amp, that actually results in more complex overtones that can sound more aggressive than just the odd order harmonics of a push-pull power amp. Then there's preamp vs power amp distortion, sag, and intermodulation distortion caused by heater hum and supply ripple. All these things make a guitar amp pretty complicated.

[@debilthomes501]

Great video. Really enjoyed it.

[@Lif-999]

Good presentation! With just a basic knowledge of the various componets, the viewer should gain a basic theoretical insight into the circuits discussed. I particularly applaud the disclaimers regarding buyer expectations and vendor snake oil claims. These things bear repeating.

[@hanksta34]

Timely. Skunkie Designs covers similar topics and you really provide some excellent tutorials to increase my understanding and knowledge..

[@Lancaster_Hi-Fi]

Awesome, thank you!

[@EddyTeetree]

Is that skunkie person a guy?

[@Lancaster_Hi-Fi]

No, gal.

[@sammys7518]

There is such a thing as too simple, SETs being an example. For instance: 1. The damping factor varies from low to zero as the signal varies and maximum power is achieved. This means the amplifier has little to no control of the speaker drivers/cones. 2. The intermodulation distortion (IMD) is roughly 3 times that of harmonic distortion (HD). So at 5% HD, one can figure 15% IMD. At 10% HD, 30% IMD etc. PP has less problems, and can include even order distortion products if properly designed, just not as much as SETs. 3. Higher order harmonics are in order due to the preceding driver stages having to furnish the large signal to drive the output tubes, 300bs etc. (The harmonics are multiples of the fundamentals. For instance the 2nd harmonic of 1khz is 2khz, the 3rd harmonic is 3khz etc. 4. Distortion producer inductors/chokes are almost always used in the signal path which creates frequency distortion (RCA Radiotron Designers Handbook, by 26 engineers). 5. Interstage transformers are often used, which creates phase shift and limited frequency response (FR) problems, just as the NOS tubes of yesteryear do, with poor specs. The main problem with Push Pull amplifiers is they are extremely difficult to properly design, and be accurate, but it can be done. Otherwise, PP amplifiers pretty much eliminate the problems of SETs listed above. A PP amplifier can actually be designed to be more accurate than any SET amplifier by definition. Very very few PP amps are accurate. Cheers

[@paulfalke6227]

Hello sammys7518, can you please give references to your numbers? As I understand you, you say that SET is bad. Is this correct?

[@sammys7518]

@@paulfalke6227 Actually some SET amps sound pretty descent, but they do have inherent limitations which limits accuracy. PP amplifiers are almost impossible to design properly, but they have the potential to be quite superior, accurate. I worked in the college lab and then established my own lab for some 40+ years now. It took specialized listening tests/testing procedures and years to actually develop a PP amplifier that was accurate. It took accurate polypropylene power supplies to reach truly natural and accurate sound. Electrolytic PS caps limit accuracy in and of themselves. Responding to point 1 of SETs. Check the output tube curves and the rp rises as the tube's plate current drops, but not in a linear fashion. Thus the damping factor drops in a non linear fashion with SETs. Pt. 2. Check the RCA Radiotron Designers Handbook, 26 engineers, 1960. Pt. 3. RCA Radiotron Designers Handbook, and Eimac for 3/2 power law. Also, harmonics add and rarely subtract from stage to stage. For instance, 2nd of driver stage and 2nd of output stage (2nd/2nd) = 4th harmonic. 2nd/3rd = 6th harmonic, 3rd/3rd = 9th harmonic. Tubes following the 3/2 power law produce 2nd, 3rd, 5th, and even 7th harmonics. Combine with previous stages produce very high harmonics, although small value. Pt. 4. Two factors present when inductor is next to the decoupling capacitor, the signal path. The inductor has DC resistance and inductance (let's forget about internal capacitances etc.). The combination causes a curved reactance vs frequency, especially at lower/mid frequencies, and definitely affects the musical quality. Best to use a non inductive resistor (better than an inductor by far) next to the decoupling capacitor. Pt. 5. Interstage/ multiple transformers have limited response. DB variations add through out the audio system and cover multiple octaves which affect sonics (Rane Corp.) Older NOS tubes have high rp and RL, plus lower gain for more stages, thus poorer high frequency response. Hope this helps Paul sam

[@ceylonmooney]

i appreciate your teaching. good video.

[@CollectingRetro]

Parts are more critical in a SET amp and differences are noticeable. It's work experimenting and trying various parts in the signal path.

[@Lancaster_Hi-Fi]

I agree! Check out my latest video, which is a short addendum to this one.

[@neps4th]

You done good, son !

[@solodolo6018]

Great video!

[@Lancaster_Hi-Fi]

Thanks!

[@royrobles7423]

Tube amps...all the way....for the win...

[@jamcdona]

I would love to see a video on hybrid amps. I just picked up a Dayton Audio unit to mess around with, and I really don't know what to think. I think I like the thing...LOL

[@Lancaster_Hi-Fi]

Hybrid can mean different things. The general usage refers to tube preamps combined with solid state power amps. The term could be extended to refer to any tube amp that includes solid state devices, whether diodes for solid state rectification or regulators for power supplies to preamp and input-stage tubes and grid bias. Solid state devices can also be used in buffers, whether input, output, or interstage. I typically use "hybrid" systems with tube power amps and solid state preamps.

[@Lancaster_Hi-Fi]

More to your point, I've never had the typical sort of hybrid amp, and I don't expect to get one. I could see adding pre-out and main-in jacks to an existing integrated amp or receiver, and I have a couple of candidates for that.

[@jamcdona]

@@Lancaster_Hi-Fi Thanks for the clarity, sir. I'm still learning- this is new stuff for me and I'm watching everything I can!

[@Lancaster_Hi-Fi]

As am I.

[@Charles12509]

Great Video, have you compared any First Watt Sit amps to tube amps?

[@Lancaster_Hi-Fi]

No.

[@brokenmechanic]

i have a question. what effect would you get if you built the chassis using copper plate.

[@FlixTraveler]

Hi, i bought a tube end amplifier. How do i know it is singele end or push-pull? It sounds good. All i know the big tubes are KT88 and it is hand build by a musician. The seller told me it's 2 x 16 watt. Can provide you some pictures if you want?

[@Lancaster_Hi-Fi]

If it's using KT88's and produces "only" 16 wpc, then it's probably single-ended. I think KT88's will produce 60 wpc in push-pull. The simplest way to tell is to count your KT88's. If you have two, then the amp is single-ended. If you have four, then it's probably push-pull.

[@WilliamEPerry]

No Iron for "best" amp? Output transformers are impedance matching devices and distort the waveform.

[@EddyTeetree]

Great presentation. Very informative. Thanks. Probably from what im seeing “audiophiles” re-discovered the SET amp designs and then hyped them as oh so special because well thats what they do. Now every shill promotes class D hahaha.

[@Lancaster_Hi-Fi]

You're welcome!

[@kevinhamming4514]

push - pull Amps have less distortion overall proven by the comparison of the 2 amps by Fred Nachbaur on dogstar music there is less THD with the pushpull compared between MiniBlock I and miniblock II, a SET and a push - pull, they use the same tube at the same operating point. A 20 minute discussion is not enough time to give adequate advice on buying an amp.

[@Daniel-79]

I know a HiFi guy that’s all about using 200 dollar fancy RCA cables. Maybe I’m cheap but I just can’t fathom how a 200 dollar pair of RCA cables could sound a 100 times better than a 2 dollar pair of cables. I’m sure they carry a signal that is pure but I don’t think my ears could easily hear the difference but my wallet sure can

[@Lancaster_Hi-Fi]

If you spent that much on a pair of cables, you'd manage to convince yourself they were worth the money! 🤣

[@williamcampbell3868]

Cables do make a worthwhile improvement in the musicality of a good hi-fi system. If you have been using cheap RCA cables for a period of time you will hear a dramatic change if you switch to more expensive cables. However the best value in audio interconnects to me is between $50 and $100 dollars. If you go higher the gains vanish and you will hear hardly no improvement if any.

[@Lancaster_Hi-Fi]

I haven't yet spent that much, but I've gone to $20-30 rather than $2. One of these days, I might go a bit higher, but at that point, I'd have to do some serious listening tests to see if the higher price was worth it. I went with the $20 cables largely for looks.

[@williamcampbell3868]

@@Lancaster_Hi-Fi Often the looks can be deceiving, especially with cables. One set will cost a few dollars more than the other and not have any performance gains. I have also seen cables that are high quality at a lower price than I mentioned before. These are in the $35-$50 dollar range.