Head-Fi.org › Forums › Equipment Forums › Sound Science › Differences between tube amps and solid-state amps - How do I know which one to get?
New Posts  All Forums:Forum Nav:

Differences between tube amps and solid-state amps - How do I know which one to get? - Page 2

post #16 of 44

Personally, I think tube amps are just a fetish object. Those who really care about sound quality use solid state amps. If I want to look at glowing tubes, I'll string up some Christmas lights.

post #17 of 44

C9 bulbs sound a bit warmer than standard C7, and the new LED units have a very cold icy sound. 

 

Seriously, the placebo and nocebo effects are real and well documented. When present, there's simply no escaping their influence.  And those effects are completely valid reasons for choosing a particular amp type.  

 

My high school economics teach taught me one thing: economy is 'maximizing enjoyment'.   Reality has very little to do with it, unless you enjoy reality.  For some of us, truth in audio has been a life-long pursuit, for others, mythology remains more fascinating.  Who's to argue? Both end results are enjoyable to their respective pursuers. 

post #18 of 44
Thread Starter 
Quote:
Originally Posted by bigshot View Post

Personally, I think tube amps are just a fetish object. Those who really care about sound quality use solid state amps. If I want to look at glowing tubes, I'll string up some Christmas lights.

I'll probably end up buying a tube amp out of novelty (Woo Audio, Schiit Audio, and Bottlehead all make cool-looking tube amps in my opinion). I haven't had the chance to listen to tube amps in great detail, but my brief impressions with them is that they do indeed produce a "warm" sound, in comparison to most solid state amps I've tried anyway.

 

Right now I own a JDS Labs Objective2, which is arguably one of the better solid state amps you can get for $144, and it sounds really good. There's nothing "special" about its sound though...it's just "transparent" and clean. I ordered an AKG K 701 as my home reference headphone and I do plan to go to a local Head-Fi meet in the near future.

 

It'll be interesting to compare the K 701 on a solid state amp and a tube amp. I believe there is going to be Woo Audio, Schiit Audio, and Bottlehead amps at the meet, conveniently enough.

post #19 of 44

Right now actually after hearing quite a few amps, including the tube ones. With low impedance phones I still say they still sound like utter crap due to noise and stuff but with high impedance cans they can actually help with produce a more desirable sound while retaining fidelity(due to weakness to actual fidelity being the can itself). 

 

Other then that, tubes are probably the only real way to high voltage fidelity like in electrostatics as of now. Till then though, my next amp and headphone purchase will probably a normal dynamic(or modded T50RP) with a SS amp.

 

On another note, I spelled discrete wrongly, do help me correct it LOL


Edited by firev1 - 12/3/12 at 6:25pm
post #20 of 44
It's all "valid" so long as everyone recognizes the presence of subjective opinion.
post #21 of 44
Quote:
Originally Posted by firev1 View Post

     Quote:

Tube sound can be described as warm, smooth, musical and other esoteric terms. Their soundstage are also typically wider due to the supposed combination of bad phase response(on scopes) and Total Harmonic Distortion. THD wise, they have significant levers of 2nd and 3rd harmonics concerting with 4th and 5th harmonic distortion to give that nice full sound.

(On scopes)

 

So is what you are saying that the soundstage of solid-state amps suffers from insufficient knowledge on the part of engineers to design past how solid-state amps can deteriorate soundstage. And if an engineer with sufficient knowledge of what test equipment to use and how to use it were to design a solid-state amp he or she could design one with all the soundstage benefits attributed to different tube amps?

 

I have spent a some time reading tube rolling threads and I'm surprised that changes in soundstage are often attributed to different tubes. Is all that attributed to different tubes selectively distorting different frequencies? Or does it have something to do with higher order harmonics levels that solid-state engineers often ignore the effects of? Can anything that tube amps can achieve be achieved by an extremely skillful engineer designing a solid-state amp? 

 

 

I am particularly interested in the answers to the questions I posed above in regard to soundstage?

post #22 of 44
Quote:
Originally Posted by bbmiller View Post
 

I have spent a some time reading tube rolling threads and I'm surprised that changes in soundstage are often attributed to different tubes. Is all that attributed to different tubes selectively distorting different frequencies? Or does it have something to do with higher order harmonics levels that solid-state engineers often ignore the effects of? Can anything that tube amps can achieve be achieved by an extremely skillful engineer designing a solid-state amp? 

 

I am particularly interested in the answers to the questions I posed above in regard to soundstage?

 

First, you have to understand that tubes are not high tolerance devices. Deviation of 5 to 10% are completely normal. Curves published in datasheet are average measurements, not something you should expect to the last %.

 

Secondly many tube amps do not use global negative feedback to linearize the amplifier.

 

Third, the parameters of a tube are heavily dependant on its operating point (voltage from cathode to anode, current). Even basic parameters such as "gain".

 

Which means that the sound  of your amplifier is dependant on a particular tube, with a particular set of gain curves. If you change the tube, you get another gain curve. But it also means that tuberolling will change the equilibrium in between channels; tubes are never perfectly matched. This will affect soundstage

 

All this will be more apparent if the tube amp is badly designed, putting the tubes in particularly non-linear sections of their operating curves. It is also more apparent if you roll tubes which are not exact equivalent, and thus have different optimal operating points. Or if you get tubes which are particularly mismatched.

 

HOWEVER:

 

Many tubes can be extremely linear if run in the right conditions. My current tube amp is an hybrid tube+mosfet amp, all the voltage gain being provided by the tube, without global feedback. Distortion at 1K, 1Vrms into 300R is only 0.028% and the frequency response is perfectly flat. Since it's only 2nd and 3rd harmonic, it's under the audibility limit. Some tubes are a lot more linear than that, over huge voltage swings.

 

 

This said, I don't think the tube amps I heard had a better soundstage than my Gilmore Dynalo solid state amp. Nor a worse one, if the tube amp was any good. I still subjectively prefer my tube amp with my hd650. It might be the tiny bit of 2nd harmonic that I get when pushing the amp at higher levels. The fact that it is much more fun to build a tube amp plays a role too. ;-)

post #23 of 44

Traditionally, the word "soundstage" is used by audiophiles as a catch all, vague term to describe a completely subjective impression. It doesn't relate to any specific aspect of sound reproduction. The word does have a meaning, but it's rarely used in that context.

post #24 of 44
Quote:
Originally Posted by 00940 View Post

Many tubes can be extremely linear if run in the right conditions. My current tube amp is an hybrid tube+mosfet amp, all the voltage gain being provided by the tube, without global feedback. Distortion at 1K, 1Vrms into 300R is only 0.028% and the frequency response is perfectly flat. Since it's only 2nd and 3rd harmonic, it's under the audibility limit. Some tubes are a lot more linear than that, over huge voltage swings.

 

 

This said, I don't think the tube amps I heard had a better soundstage than my Gilmore Dynalo solid state amp. Nor a worse one, if the tube amp was any good. I still subjectively prefer my tube amp with my hd650. It might be the tiny bit of 2nd harmonic that I get when pushing the amp at higher levels. The fact that it is much more fun to build a tube amp plays a role too. ;-)

Quote:
Originally Posted by tomb View Post

 

Switching power supplies are often very noisy.  They claim to have "pushed" the noise well above the audible threshold, but there are many artifacts, phase distortions, etc., that cause harmonics into the audible band.  Let me emphasize, however, that "noise" in the power supply is rarely something that's heard.  Instead, it's a general lack of quality in sound: maybe a glare in the high-end, a general loss of dynamics, sloppy bass, etc.  It simply causes effects that result in sounding blah vs. GREAT.  Only extreme examples - not worthy of any consideration at all - result in noise that's directly audible in the sound signal.

 

Also, switching power supplies can sometimes be used with amplifiers (including headphone amplifiers) with good success.  However, every amplifier circuit has a calculated property known as PSRR - Power Supply Rejection Ratio.  It's a measure of how much noise in a power supply will affect the amplifier circuit itself.  Good circuits have very high PSRR's.  However, powering a source such as a DAC does not qualify.  Once noise is introduced, it'll propagate throughout the signal stream.

 

In regard to the two above quotes I have these questions in the first quote the respondent mentions his tube amplifier has very low distortion operating in the linear range without feedback to improve the distortion. In the second quote the respondent mentions in audible switching power supply noise being out of the audible spectrum yet affecting sound inside the audio spectrum. So combiningthe two ideas could achieving low distortion and solid-state amplifiers by feedback be deterius of sound in the audio spectrum even if all the bad affects of feedback reduction in distortion occurs outside the audio spectrum? And are there solid-state amps that find solid-state devices which operate with low distortion because they operate in the linear portion of these solid-state devices like the first respondent's tube amplifier?

 

 

I am also wondering when the second respondents mentions PSRR do tube amps generally have a very high PSRR? I haven't heard any people mentioning linearity of power supplies in regard to tube amps. Is it just not a issue with tube amps?

post #25 of 44
"general lack of quality in sound: maybe a glare in the high-end, a general loss of dynamics, sloppy bass, etc"
 
This seems like the sort of issue that crops up everywhere except it mysteriously disappears in blind tests.

 

Properly designed solid-state amps operate in the linear region. In fact this is easier to do with solid state components than with tubes.

post #26 of 44

The best tube amps sound almost as good as solid state.

post #27 of 44

Probably the knowledge, especially with regards to which distortions weigh more on the human hearing, even now its not really known. Soundstage IME is a very real thing though, imo phase and harmonic distribution has the most to do with it along with other factors.

 

I tend to think that some designers would focus on a few electrical parameters they would want to optimise and see whether it sticks in the subjective listening experience as there is a lack of correlation between the conventional objective data and the subjective listening experience. 

 

Also compared to transducer distortion, one has to note that amplifier/dac distortion is considered marginal.

 

As for PSRR, tubes do not have typically high PSRR as SS circuits though some designers will go the extra mile in designing a good power supply. Also some SS amplifier circuits may have poorer PSRR in favor of other electrical parameters.

post #28 of 44

Actually I have more questions as I'm still very much a noob about tube amps (BTW I have at most extremely basic knowledge of electronics)  :-

1) OTL Amps - Output transformerless

                           Known for high output impedance but how high? The Zana Deux is 3-12 ohm.A recent post mentioned 120 ohm Zout on average.

                           a) Aside from repetitive posts on suitability to HD800, are there any other headphones normal headphone OTL Amps are suited for?

                           b) I read somewhere that early OTL designs, the driver tubes had potential to explode. Is there no concern for modern OTL amps now?

2) SET Amp - Single Ended Triode...but does that mean it it must have a transformer in the output stage? i.e. are SET and OTL mutually exclusive?

                         How is it different from the other amps? By difference I'm curious on the following :-

                         a) design

                         b) practical implications (e.g. high output impedance?)

                         c) special sound characteristics?

                         d) special way of maintaining or using it?

3) Push Pull - What's the difference with this vs SET Amps and OTL amps? By difference I'm curious on the following :-

                          a) design

                          b) practical implications?

                          c) special sound characteristics?

                          d) special way of maintaining or using it?

3) Hybrid - which part of the amp is hybrid? Where are tubes normally used in this case? In the driver stage? Input stage? Rectifier stage?

4) OCL Amp - Ouput Capacitorless amps.

                          a) does anyone make these for heaphones? Are there any examples?

                          b) Same questions as above, what are the practical implications, special sound characteristics, and special way or using it?

post #29 of 44
Quote:
Originally Posted by bbmiller View Post
 

In regard to the two above quotes I have these questions in the first quote the respondent mentions his tube amplifier has very low distortion operating in the linear range without feedback to improve the distortion. In the second quote the respondent mentions in audible switching power supply noise being out of the audible spectrum yet affecting sound inside the audio spectrum. So combiningthe two ideas could achieving low distortion and solid-state amplifiers by feedback be deterius of sound in the audio spectrum even if all the bad affects of feedback reduction in distortion occurs outside the audio spectrum? And are there solid-state amps that find solid-state devices which operate with low distortion because they operate in the linear portion of these solid-state devices like the first respondent's tube amplifier?

 

I am also wondering when the second respondents mentions PSRR do tube amps generally have a very high PSRR? I haven't heard any people mentioning linearity of power supplies in regard to tube amps. Is it just not a issue with tube amps?

 

I don't think you can associate the two ideas that way but to answer some questions:

 

 - Yes, with jfet or mosfet, one could have a solid state amplifier without global feedback. You can find such things with Nelson Pass preamp for example. It's however harder than with tubes to get real low distortion. With transistors,it's harder, as transistors really aren't very linear. In either case, you'll need local feedback.

 

- Now, global negative feedback isn't a bad thing at all if done properly. There aren't bad effects to it. If you want, there's an excellent article by Bruno Putzeys about that: http://edn.com/design/consumer/4418798/Negative-feedback-in-audio-amplifiers--Why-there-is-no-such-thing-as-too-much 

 

- PSRR in tube amps heavily depends on the topology used. It can be made very good. Or very bad. If you go to diyaudio, you'll find a lot of discussions about power supplies for tubes, regulation etc.

 

 

NB: all this is gross oversimplification.

post #30 of 44
Quote:
Originally Posted by AnakChan View Post
 

Actually I have more questions as I'm still very much a noob about tube amps (BTW I have at most extremely basic knowledge of electronics)  :-

1) OTL Amps - Output transformerless

                           Known for high output impedance but how high? The Zana Deux is 3-12 ohm.A recent post mentioned 120 ohm Zout on average.

                           a) Aside from repetitive posts on suitability to HD800, are there any other headphones normal headphone OTL Amps are suited for?

                           b) I read somewhere that early OTL designs, the driver tubes had potential to explode. Is there no concern for modern OTL amps now?

2) SET Amp - Single Ended Triode...but does that mean it it must have a transformer in the output stage? i.e. are SET and OTL mutually exclusive?

                         How is it different from the other amps? By difference I'm curious on the following :-

                         a) design

                         b) practical implications (e.g. high output impedance?)

                         c) special sound characteristics?

                         d) special way of maintaining or using it?

3) Push Pull - What's the difference with this vs SET Amps and OTL amps? By difference I'm curious on the following :-

                          a) design

                          b) practical implications?

                          c) special sound characteristics?

                          d) special way of maintaining or using it?

3) Hybrid - which part of the amp is hybrid? Where are tubes normally used in this case? In the driver stage? Input stage? Rectifier stage?

4) OCL Amp - Ouput Capacitorless amps.

                          a) does anyone make these for heaphones? Are there any examples?

                          b) Same questions as above, what are the practical implications, special sound characteristics, and special way or using it?

 

1. Output impedance will vary a lot. Without going crazy, a White cathode follower made with 6AS7 will have a 70ohms output impedance, which can be further reduced with feedback to under 10ohms. But many amps will have an output impedance going up to 120ohms indeed. Output impedance is only a factor though. OTL circuits are often limited in the current they can deliver, which restrict their use to efficient headphones.

 

a/ HD650 isn't that bad too. The beyer headphones were really made for 120ohms output, so they would be fine.

b/ Not an issue today.

 

2. SE and OTL aren't mutually exclusive. In the speakers world, SET is associated with transformer coupled amps, as there isn't another way to get enough power. With the low power required by headphones, it isn't the case.

 

a/ have a look at the "A Single-Ended OTL Amplifier for Dynamic Headphones" by Van Waarde on the web. Or "A Low-Voltage Class-A Tube Headphone Amplifier" by Ahammer for OTL.

At this article http://diy.ecpaudio.com/p/parafeed-tutorial.html for parafeed (transformer coupled).

For single feed (transformer coupled), it's just like for speakers, only scaled down and with special iron.

b/ It depends if you use OTL or not.. if you do, output impedance can be quite high (but put enough tubes in // and it will go down fast). If using transformers, you can easily get around 10ohms.

c/ impossible to say. Typical feedback-less SET have highish H2 distortion. But once you add feedback...

d/ none.

 

3. Again, you can have push-pull either OTL (the Bijou for example) or transformer coupled. Bottom line is that push-pull gets you more power but the distortion spectrum isn't as nice as SE.

 

a/ have a look at the "bijou" for an example of OTL push-pull. This : http://diyaudioprojects.com/Tubes/6DJ8-Tube-Headphone-Amp/ illustrates a push-pull transformer coupled amp.

b/ usually lower output impedance (a bit).

c/ as an overly general rule, the fact that you cancel a lot of second harmonic will get you a "more dynamic sound", less "round".

d/ none.

 

4. Usually, hybrids will use a tube for gain stage and a ss stage for current buffering.

 

5. OCL and OTL would be crazy for headphones and I don't think any manufacturer is crazy enough. OCL with a transformer means a single-feed amp. I think the DNA sonett would be an example.

 

 

 

As usual, the best answer for all questions is : "it depends".

New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Sound Science
Head-Fi.org › Forums › Equipment Forums › Sound Science › Differences between tube amps and solid-state amps - How do I know which one to get?