Tube or solid state?

[outlander]

The Rotel forum might be the wrong place to ask this question but I know a lot of you probably have educated opinions on this subject, so here it goes. What do think the advantages are of a tube amp over a solid state and vise versa. The reason I ask is a friend just picked up a Jolida tube amp and he swears the improvement over his Luxman solid state is huge. But when I compare the specs of his unit to my RB-985-MKII, the Rotel has it all over the Jolida. I’ll be heading over to his place this weekend and I’ll see for myself.
O

[mjb]

I personally think tube amps CAN sound wonderful - you must ignore the specs, and listen. The specs don't usually look very good, THD for example is usually very poor, but the sound is, well, warm. There is a difference between tubes, tube manufacterers (of the same tube type), and most importantly, the output transformers (quality output transformers, required in valve power amps, can be very expensive and heavy).

But, you can't beat solid state when it comes to power and abuse at low cost. :T

A good compromise is a hybrid setup: a valve preamp coupled with a solid state power amp. The pre-amp doesn't require an expensive output transformer due to the high impeadance between stages, and you get that characteristic valve sound.

The only problem with this is I want 7 channels and a surround processor with digital input, and a remote control. Its easier to buy a quality processor/pre-amp, and they can sound every bit as good - though perhaps without that special something (harmonic distortion :B ).

[Kevin P]

I'm moving this thread over to Audio Hideout which is a more appropriate place for this discussion (since it has nothing to do with Rotel per se).

You can't really compare tube amps to solid state by reading the specs numbers, as any SS amp will walk over a tube amp in that department (e.g. power output, THD, etc.). Where tube amps shine is in their (perceived) sound. They don't need as much power to produce a sweet, warm sound that SS amps rarely achieve.

One popular theory about tube amps sounding better is because their harmonic distortion tends to be even order rather than odd order which is more "pleasing" to the ears. Thus, when a tube amp distorts, it is a "pleasant" distortion that isn't as harsh as a SS amp's distortion.

Tube amps tend to have a very sweet sounding midrange as well. I have a tube headphone amp in my "cubicle tubicle" setup at work and it sounds wonderful. Tube amps generally won't cut it in a home theater though where you need 5 to 7 channels powerful enough to reproduce the impact of a movie soundtrack. SS amps are best for that application.

[Victor]

What do think the advantages are of a tube amp over a solid state and vise versa. O

Tube-based amplifiers offer absolutely nothing. That is to say that they offer no advantages over the solid state amps regardless of the issues discussed, -be it sound quality or anything else. Anything that the tube amp can do in the sound quality department can also be done with a transistor amplifier. Carver had shown it conclusively in the mid-80s.

So if it is the warm sound that you are looking for then there is a way to achieve it with a more modern design without tubes. For instance products designed by Nelson Pass have very similar characteristics to the tube-based designs.

However, you can turn any Solid State amplifier into a “tube-sounding” amplifier for about $10. All it takes is an appropriate power resistor attached to the amplifier output and that will render the change in damping factor with corresponding increase in THD. In Blind Tests I have observed no statistical difference between a modified Bryston 4B with a resistor and $7000 Cary triode amp. The same results were observed with an older Rotel. Another contender if memory serves was a Japanese megabuck tube amp, - the Ongaku. Again, no statistical differences were heard. Resistors used were from Radio Shack, 20Watt, 8-Ohm non-inductive with several configured in parallel.

[Patt]

Tube-based amplifiers offer absolutely nothing.

That is a pretty bold statement Buddy.

What if a person, other than yourself enjoyed the sound better than solidstate or wanted to venture into something a little different. Maybe it is just visually more appealing to someone else.

[soundhound]

Little off topic, but here goes anyway.Carver proved conclusively he could design an amplifier that was in dyer need of component mounted heat sinks, or fans. I just repaired a TFM 24 for a friend, as well as 4 that I had previously owned. Every TFM series I have ever seen by carver has a tendency to burn the circuit board under the driver stages. Resistors lose their color codes, and solder joints take a beating from the heat. I touched up about 30 bad joints on my friends 24, then just for shits and giggles ran one of my fronts from 1 channel of it, and the other from my 1080. Very BRITTLE sounding, my 1080 was much warmer. I asked my wife setting on the sofa next to me what she thought, she said "I like it, it's crisp sounding". This coming from a woman who is content with a G.E. boombox in the kitchen....... Carvers are great for people who want to pound the heck out of a set of cerwin vega 18's, but are not my prefered choice to kick back and enjoy some tasty music with.

[Victor]

That is a pretty bold statement Buddy.

What if a person, other than yourself enjoyed the sound better than solidstate or wanted to venture into something a little different. Maybe it is just visually more appealing to someone else.

Indeed, what I said might be construed as somewhat too forward and I am sorry if I offended you. However, I never implied in my post that one cannot enjoy using tube-based amplifier. If the ‘tubes’ do it for you, - then great! I would never presume to tell anyone what to do.

My point was simply that the tube-based electronics has served it purpose over time and at present it no longer offers any benefits at all. It is an opinion that I can support, but I do not tell you what choice to make in your selection of commercially availably amplifiers, - tube-based or otherwise.

I do not dispute that the tube-based amp with its commonly used single-ended output circuit topology would generate even order harmonics with corresponding pleasing effect on human ears. This is a fact. It is also a fact that those harmonics constitute what we commonly call distortion. In the case of a tube amplifier that distortion is on the order of 1-2%, which you would agree is horrendously high. It is not to say that one cannot design a tube-base amp with much lower distortion, but then it will destroy the charm of the tube amplifier and its claim to fame. And so the game goes on, - tubes and distortion purposely go hand in hand.

The question is, - do you want this distortion or not? If you do, then my point is simple, - I can produce this distortion ahead of an amplifier with a low-power circuit that will cost about $10 in parts or with a signal processor that will cost about $200 or with an output resistor at the tail end of any solid state amplifier for even less money. There is no magic here and no need to spend megabucks to achieve the sonic nirvana of purposely distorted sound.

In principle and by definition, amplifiers are Linear-Time Invariant Systems and, as such, they are not supposed to produce any distortion at all. Amplifiers simply change the amplitude of the input signal while faithfully preserve the signal’s spectral content. If the amplifier changes the signal frequency response or introduce unwarranted distortion then we should not call an amplifier, - we should call it a Power Tone Control. Naturally zero distortion is not possible, but in human terms we can get very close, - look at Bryston amps with 0.007% THD across their product line.

..regards

[Victor]

Little off topic, but here goes anyway.Carver proved conclusively he could design an amplifier that was in dyer need of component mounted heat sinks, or fans. I just repaired a TFM 24 for a friend, as well as 4 that I had previously owned. Every TFM series I have ever seen by carver has a tendency to burn the circuit board under the driver stages. Resistors lose their color codes, and solder joints take a beating from the heat. I touched up about 30 bad joints on my friends 24, then just for shits and giggles ran one of my fronts from 1 channel of it, and the other from my 1080. Very BRITTLE sounding, my 1080 was much warmer. I asked my wife setting on the sofa next to me what she thought, she said "I like it, it's crisp sounding". This coming from a woman who is content with a G.E. boombox in the kitchen....... Carvers are great for people who want to pound the heck out of a set of cerwin vega 18's, but are not my prefered choice to kick back and enjoy some tasty music with.

I know that early TFM amps that Carver offered had issues with reliability. However, those amps still had shown that the transfer function modification will result in a perceived change in sound, - and that was the point I was trying to make.

Also, I was not thinking about the TFM amp, - I was thinking about a more modern offering from Carver, - the Sunfire. As you probably know, the Sunfire amp did offer a ‘high-impedance output’. This output had a resistor attached to force the output resistance into the stratosphere. Carver did this on purpose in order to attract the tubeophiles. Although it makes very little sense from electronics perspective, it did generate more sales. Bottom line is, - any amp can be ‘voiced’ to sound like anything you want; - you do not need tubes and corresponding expense to do it.

The reliability of Carver products have improved over the years and now it is on par with anything else out there, so the problems of TFM amp are now an ancient history and can be thought of as an anomaly associated with a start-up nature of the Carver’s business 25 years ago.

[Nick M]

I never liked the tube amps I heard. Sitting in the "Big-Boy Room" at my local hifi shop swapping back and forth between a Krell and a Conrad tube amp, it was quite easy to choose which was which. The tube amp definetly gave the speakers a warm sound. Combined with the fact that these were big expensive warm sounding B&W units, it was exacerbated even more and sounded borderline fuzzy.

To be honest, I can hear the difference between tube vs. solid, and receiver vs. seperates, but hearing the difference between a Rotel RMB-1095 and a Carey Home Theater amp is beyond my ear's capabilities. Either that or I need to use more familiar speakers and surroundings.

[Kevin P]

According to this thread, tube amps are not recommended for B&W speakers (specifically the 703s). So don't discount tube amps entirely if you weren't impressed with the sound with a particular set of speakers, especially ones that are on the more difficult-to-drive side like B&Ws.

Most tube amps have trouble with low impedance speakers or ones whose impedance fluctuate widely at different frequencies. High efficiency speakers with relatively flat impedance curves tend to work better with tube amps in general.

Tubes produce voltage, and can produce a lot of voltage, but they produce relatively little current (though the use of transformers mitigate this somewhat). SS amps produce current, and lots of it. Thus, SS amps are better for driving low impedance loads that draw current, and tube amps shine with higher impedance loads that thrive on higher voltages.

[Patt]

Hi outlander,

That is pretty neat being able to check out equipment like that at a friends house. I would be interested to hear what you find out about the different pieces of gear.

[outlander]

Thanks for the replies guys. I’ll let you know what my impressions are of my first tube amp listening experience after I check it out (hopefully this weekend). It will be an interesting comparison to my Rotel RB-985MKII considering we’re both running Polk Monitor 10B’s for speakers.
O

[outlander]

Here they are, my impressions of my first tube amp listening experience. I’ll keep it short and to the point. Mid range frequencies were beautiful, very smooth and natural. Guitars and vocals really sounded nice. The highs were clear, maybe slightly brighter than a solid state amp but not offensive. The problem in my opinion was in the low frequency range. The low end punch of his system was gone, in fact there was very little bass at all. Without the low frequencies it tended to leave the music sounding a little flat. But even with that said the detail and imaging was excellent. I think what it comes down to is the tube sound is an acquired taste. To sum it up I would say the detail in the mid frequencies has it over the solid state but this slight improvement is not worth the big time loss you take in the bass.
O

[Patt]

Thanks a lot Outlander.....I've heard the same about good midrange and possible lack of bass. Maybe it is system dependent.?

[stantheman2]

A couple thoughts...

Re the previous post, tube amps typically have a lower damping factor than sold state amps, which is perceived as "slow" bass and a lack of bass "authority" - i.e, the bass doesn't sound as good.

Re Victor's posts, he seems to focus on low THD as the prime indicator of sound quality. IMHO, it just isn't that simple. An amp designer could crank in an insane amount of negative feedback so that the amplifier, when tested with a frequency sweep of simple sine waves (the standard test procedure), would produce infinitessimal THD. However, most audiophiles would agree that such an amp wouldn't sound very good when reproducing complex musical signals. Rotel has a good rep on this site for producing great sounding amps (for the money), but their rated THD is several times higher than some (much cheaper) Sony gear. THD, IMD, damping factor, headroom, ability to produce higher power into lower impedances, stability and control into complex loads, etc, all affect how an amplifier will sound. Tubes and transistors each have their inherent advantages and disadvantages, but a transistor-based solution would typically be cheaper and have a longer service life for a given rated power output, and hence transistor amps have a much higher market share than tube amps.

[Victor]

A couple thoughts...
Re Victor's posts, he seems to focus on low THD as the prime indicator of sound quality. IMHO, it just isn't that simple. An amp designer could crank in an insane amount of negative feedback so that the amplifier, when tested with a frequency sweep of simple sine waves (the standard test procedure), would produce infinitessimal THD. However, most audiophiles would agree that such an amp wouldn't sound very good when reproducing complex musical signals.

You can’t just crank out the insane amount of negative feedback and still have a reasonable close loop gain. Normally, most amplifiers in both camps feature about 30 dB of close loop gain. Negative feedback is used to flatten out the frequency response of the amplifier. In other words, the designer trades gain for speed or frequency through the use of negative feedback.

In order for the negative feedback to work properly the amplifier must have a suitable open loop gain at the frequencies of interest. In the case of most tube amps the open loop gain at DC is about 50-60 dB. It is more like 80-120 dB with most well designed transistor-based amplifiers. This value drops off at 20 dB every decade, courtesy of the compensation network used to preserve an overall stability of the amplifier.

Well, let’s see where does this math takes us, - with 40-50 dB of open loop gain at low frequencies we get a unity gain frequency of about 50 kHz, maybe even 100 kHz. Definitely at 20 kHz the open loop gain is only about 35-40 dB or thereabout. Once we close the loop to set the overall response, we only get about 5-10 dB of excess gain. This is the gain that is responsible for the overall functionality of the negative feedback stracture and is the reason why many tube amps have a very poor THD performance.

It seems that at higher frequencies the tube amps function with very little to nonexistent negative feedback with consequential luck of control. In order to regain this control we must lower the close loop gain of the amplifier, but then we will end up with an amp that has 15-20 dB of voltage gain and even that on a good day. So in the end it is not that simple to simply increase the negative feedback of a tube amplifier in order to improve the THD numbers.

Testing with the sine waves is perfectly acceptable practice. Fourier analysis tells us that a sine wave is the most primitive signal you can possibly have and as such it is a fundamental function that is at the core of all other function. Although music is seen as a stochastic or non-deterministic process, still Fourier applies. Therefore, in principle any musical passage can be represented by a bunch of sinusoidal signals of different amplitudes and harmonically related frequencies. This is the reason why it is fine to use sine waves for testing, since the music is really complex combination of those sine waves.

Tube amps usually have a rather pour dampling factor. In fact most tube amps have no damping factor to speak off above 10 kHz. In other words the output impedance of the tube amps goes into stratosphere at that point. This impedance forms a low-pass filter with the capacitance of the speaker cable particularly if the audiophile in question uses an exotic cable with habitually high capacitance. This low-pass effect is clearly audible.

In any event those are just a few problems that tube amps are suffering from. There are many more. I would also say that it is possibly to design a tube amp that will have none of these problems, but then it will be a fine example of engineering and it will sound just like any solid state amplifier, - that is to say it will have no discernable sound. The question is why bother? May as well buy a solid state amp to begin with.

[autio]

I really am not prepared to chime in on the tube ss debate but I will mention this (since Damping factor has been mentioned many time in this thread) that there is some debate to the importance put on this spec. Check out these links

http://www.audioholics.com/techtips/audioprinciples/amplifiers/dampingfactor2.php

http://pssaudio.com/english/art010.htm

http://www.classic-audio.com/marantz/mdampingfactor.html

[autio]

Victor,
you seem to know what you are talking about more than I so why dont you put things into perspective, when talking about tube amps and damping factor you said

Tube amps usually have a rather pour dampling factor. In fact most tube amps have no damping factor to speak off above 10 kHz. In other words the output impedance of the tube amps goes into stratosphere at that point.
(damn I wish i knew how to quote someone the right way)

But in doing a little reseach I found that Crown (one company that believes in high damping factors) thinks damping factor at the freq you mention is not all that important

"We have chosen to start using a frequency range for testing that is more applicable to were Damping is the most critical and hope that other manufactures will also follow suit. Some manufactures will even give their Damping spec. at 1Khz, which is in a range, were the mass of the driver doesn't need that much control. These types of specs. basically make them look good. Other manufactures don't even tell you what criteria they’ve used for deriving their Damping number." (taken from crown forum)

so that being said explain how damping factor at 10khz is important to the tube ss debate ?

[Victor]

Autio,

We can discuss the importance of the Damping factor at length, and it has been done many times before. In fact, the links you provided do explain this topic in significant detail. One point, however, needs to be stressed; - the high damping factor is a good thing. We may wonder exactly how important that is, but regardless, the high damping factor can only make things better, although marginally so.

In my opinion, an amplifier that sports a high damping factor is indicative of a good overall design. I understand that at high frequencies the damping factor is less important because the mass of the transducer is rather low and we do not have inertia issues of any significant magnitude. Still a low damping factor indicates a rather high output impedance and that in itself is an indication of other problems that may exist with the design. One of the problems that immediately come to mind is this, - if an amplifier has a rather low open loop gain, as in the case of tube amps, that would lead to insufficient negative feedback which would in turn lead to higher output impedance with consequent low damping factor and higher distortion.

Naturally we can say, well, we do not care if the output impedance is high because its effect on the speaker control may be negligible. Sure, but it also leads to increased distortion. I guess my point is that you cannot look at a system such as a power amp and isolate a particular parameter in exclusion of other parameters. A well designed amplifier must have low distortion across the power band and that in itself would force the damping factor to also be quite low.

[scottielee]

no doubt solid state can sound like tubes. but i haven't heard solid state gear that sounds the same or better than top tube gear (lamm, jadis, canary, bat). i may not have heard the best solid state yet, but it seems to me that great tube gear usually cost less than equal sounding solid state. plus tubes just looks cool.

[DifferentLee]

I tried a number of high end SS amps and hybrids and was disappointed. I eventually bought an Audio Research VT-100 amp which works incredibly well with my Maggies. I find tubes to offer up a more natural midrange than all but the most pricey SS amps and for me midrange is critical. Modern tube amps from ARC and C-J can do bass just as well as SS as well which contradicts a well-worn myth. In fact, the mid-bass is often spectacular.

One must also bear in mind that certain speakers mate better with certain amps. Component matching is often very important.

[georgev]

Take the time to read this article. V. disturbing and ?enlightening.

http://www.theaudiocritic.com/downloads/article_1.pdf

It is titled the 10 biggest lies in audio. Deals also with tubes/vs. SS.

If the link does not work, go to the audiocritic.com then to sample articles and the click on the article.
Be keen to hear your thoughts.

[Patt]

Sure, everyone is entitled to their own opinon.

[georgev]

Spot on Pat, which is why I want to try tubes in my system at the pre amp stage.
Currently all transistors. Sounds pretty good as is but I am told that I must try the tubes.

[Patt]

You know...It crossed my mind about a tube pre to go with my new (used) CJ amp and try it with my Klipsch 4.2's in the back room. There is something that peaks my curiosity about the whole tube thing.
I've read where one guy said the tube pre is an in-expensive way to try it to see if it floats your boat before going the whole tube pre/amp combo.

May just very well be a cool different way to achieve the sound a guys looking for, rather than just the regular old reciever speaker combination.

[scottielee]

hi georgev,
interesting article. while reading, i didn't doubt for a moment what i have learned through past experiences with different solid state/tube gear and power conditioners. i have never done a/b/x tests, but i have heard and seen very obvious differences between cables and conditioners on audio and video. i have even heard the effect of a harmonix cd mat (not jaw dropping but noticeable) at the past 2005 nyc show. i don't have golden ears or eagle eyes, but i honestly try very hard not to lie/make up findings when auditioning gear.
on ss vs. tubes, i agree that ss can sound better than tubes. it is just harder to find great sounding ss gear that cost less than tube counterparts, in my experience. does that mean ss manufacturers are asking for bigger profits to pay off their r&d, whereas less tube r&d are required nowadays considering tubes' longer history?
anyway, thanks for the article. stuff like this keeps this hobby fascinating.
scottie

[DifferentLee]

That article is the usual hoohaa from Peter Aczel. He still can't accept that DBTs are not the end of all component assessment, but to claim tubes are inferior to solid state for audio is just beyond me. He clearly has not heard what the best stuff is capable of.

[David Meek]

Friendly reminder: Let's keep this on the intersting track it's currently on, and not turn it into a hack on reviewers. Thanks.

David

[georgev]

I run an all Levinson system: front end, preamp, power amp into Nautilus 802's.
Should I audition tubes at the preamp stage and if so which would you guys suggest.

[David Meek]

George, since you are already running high-end (Mark Levinson :drool: ), I'd suggest something along the lines of upper end Conrad-Johnson or BAT (Balanced Audio Technologies).

[georgev]

David, I had thought of the BAT. Would you consider auditioning a tube within the setup described?

[David Meek]

I think experimenting with various combinations of equipment is a good thing, so yes, I'd audition something if it were of comparable quality. The cool thing about all this is once in a while - maybe - you can "catch lightning in a bottle" and discover an unexpected synergy that raises the level of your enjoyment. Go for it! :T

[DifferentLee]

Try Audio Research at the preamp and amp end of the system.

[twitch54]

Tube or SS ????, easy go with quality Class A SS, best of both worlds !!!!

I'm presently using a Plinius SA-102 (recomended to me buy a member of another audio site) and couldn't be happier.

[Alaric]

McIntosh amps tend to sound "valve-like" even without a tube pre-amp. Or If you don't need a lot of power , my little Marantz PM7200 puts out 25 wpc of Class A that has a very "tubey" sound. My speakers are pretty efficient , that helps.