I was a Hi-Fi snob. But some time ago I realized I was listening more and enjoying it less. I envied people who had what I considered crap systems, but who got great enjoyment out of listening. I had my moment of truth; I was chasing my tail, I wasn't listening to MUSIC anymore, I was listening to THE EQUIPMENT! That's two different things. The whole point is the enjoyment of music. It's one thing if your equipment is defective, but does one need to spend$5,000, $10,000, $20,000 or more before they can start enjoying music? Don't get me wrong, of course I would love to have that $50,000+ system with a room to match. I'm just saying, don't get lost, and don't miss the point of all this, which is THE ENJOYMENT OF MUSIC.
MrMinathecat You are so right. In fact I have been an audiophile for over 35 years and have always noticed that most audiophiles listen to music that sounds good on their systems and not necessarily the music they like best.
I also used to play with audiophile systems but not anymore. It's good to listen to that excellent systems but I found myself enjoying various music not recorded up to audiophile or even HiFi standard. I just enjoy more music than the quality of the reproduced sound. If it's good music, you can feel it's good music, no matter how good it was recorded and reproduced out of the speakers. The players have put their souls in the piece already and the audience can absorb the messages musicians wanted to convey. Most musicians I know do not care much about their listening systems. They usually use whatever they can find to enjoy music and they can still distinguish the good, the mediocre, and the bad pieces meticulously. I still have my hi-end system but I'm not playing around with it anymore. Just consider it as an excellent system to play recorded music sometimes I want to enjoy a better quality.
Personally, (just my opinion) a $5,000 - $10,000 hifi system will give you some very good musical experience, regardless of genre. Spending more will give you some improvement, but not much.
I've been waiting 25 years for that, lol. I really appreciate it. That was the simplest, most concise explanation of valves vs solid state I've heard. To the guy who dosent understand how it effects sound: remember that voltage = PSI (roughly) running at higher voltages creates a higher share of linearity for the signal. The more linearity, the more headroom. More headroom allows for broader soundstage (x&y), faster reactions to peaks and lows, it adds some warmth (although a good SS amp that's good and warm and running in the meat of it's powerband is hard to beat) and definitely "puts less strain" on the signal b/c of the exponentiated section of linearity on offer to the signal.
Thanks for posting this Paul. Much appreciated. Both you and I are men of a certain age. I am a vacuum-tube enthusiast. I have a few vintage-1962 Dynaco ST-70's and a few Dynaco PAS-3 preamps that I've kept alive for 40+ years. Thanks again for your expert explanation of tubes vs. transistors.
"Something nostalgic about that warm glow makes me feel good." === I was packed off to boarding school in the late 1960s and took my Dad's cast-off tube audio gear with me. A number of my music-loving classmates had similar stuff. All our parents had the latest shiny transistor equipment but it didn't sound so good to us. Over the years I have had some really nice solid state gear but 25 years ago I built my first single triode power amp. It felt like "coming home" and is still my first choice, despite the more costly solid state gear I have lying around. Built to satisfy rather than impress...
@@TheRealWindlePoons Does this satisfaction lies in the ritual, the tradition or anything else that has more to do with some social aspect than the music itself?
@@mignotmaxime2409 No. Not Ritual. Not tradition. Not social. The single triode amplification stage has a natural presentation I have seldom found elsewhere. But its ALWAYS about the music. Otherwise, what's the point?
@@TheRealWindlePoons Thanks. I just read the 10 biggest lies in HiFi by Peter Aczel who says tubes do not sound different. I was wondering if the preference for tubes could be elsewhere than the music. Not necessarly a bad thing but it is the case, it must be clearly stated.
Thanks for this. I just came across two young tube enthusiasts in my neighbourhood building their own preamps. I'm an RF/CE engineer and was heartened to see young people getting "hands-on" in this age of buy and throw away. I believe the sound difference is in the even octaves only that is found in tubes, whereas the transistors let thru both odd and even harmonics. There's a music theory that says even harmonics makes a pleasant sound, why some music tunes are more appealing than others.
Second, fourth, sixth and eighth harmonics are all musically harmonious. But with odd harmonics, only the third qualifies. The fifth, seventh, etc., do not form accurate musical intervals, so they cause dissonance. In addition, distortion from solid state designs tends to produce very high-order harmonics, causing a raspy, crackly or buzzing effect.
Yes, they sound different, but "different" doesn't necessarily mean better. In fact the coloration added by Tubes (pleasant or not) is often referred to as "good distortion" Tubes are used in the recording studio too. Some microphones use Tubes to warm the sound. And Jazz & Blues guitarist use Tubes for the sound they produce when over driven. I'm going to kick my own ass for saying this, but Tubes can make certain digital recordings sound better to our ears, BUT it greatly depends on what's being recorded.
Wrong ..............tubes are not warm sounding , only if not correctly implemented or wrongly used, clearly you don't know about vacuum -tube technology and that's a fact ............
Frank Geeraerts We use Tube mics in the studio to to warm or smooth an instrument. The terminology is just geek speak as long as its understood during sessions. Vocals can sound very nice with Tube pre's ... The Hi-Fi world is rather different. I wouldn't use Tubes in my system as they're not needed. And Frank, think about how you reply 😉
Frank, tube amps MAY sound "warmer" if they're single-ended design, in which case you get a noticeable portion of the 2-nd harmonic (over -60dB), which does sound somewhat pleasant. You don't get that effect from a push-pull design.
Two reasons tubes sound different, one is soft saturation, tubes can operate into a long saturation curve, two is that output is "almost" always through a transformer with hysteriesis distortion. When newer fet amps are designed to sound like a tube amp these "distortions" are programmed into the design.
This whole thing cracks me up. I'm a guitar player. I've used my old Peavey guitar amp on gigs and had other guitar players come up to me and say "... man, that thing sounds great.... that must be a tube amplifier... " I like this because I'm using a relatively inexpensive amp, that has no tubes in it. In fact, the whole amp (which is all discrete, no ICs) has only one FET in it. The rest of that transistors are bipolar junction types (BJT transistors.) Peavey was cleaver in that they did use an audio transformer on the output, which helps to add just a bit of tube like "mud" into the sound. This amp was a cheap amp, to begin with (and I've replaced some parts with parts from a surplus store, occasionally) and that's what makes is especially amusing when guitar players mistake this for a tube amp. After I reveal that this is a solid state amp, I say to them: If tubes, alone, made for audiophile level listening, every table radio from the 30s, 40s and 50s would be audiophile quality... and we know that ain't true.
cool, I always liked the way old vacuum tube radios sounded when I was a kid. As an kid/teenager during the 1960's We guys use to hangout and go everywhere with our little 9 volt transistors. Happy to hear that the vacuum tube has not gone the way of the horse.
Audio Research went back and forth between tubes and transistors in their amps and preamps and their tube designs were always better. Warmest sounding amp I have is a transistor amp made in Norway but it gets hot as a stove. Tubes don't really have a warm sound. But what they do have is a lack of that hard edge sound of alot of transistor amps have.Everything colors sound but tubes tend to sound more like the real thing.
Class A amplification always sounds better, even if it is solid state. JVC had some really good amps that were super cheap (less than $100) , but they had poor speakers. I suppose they quit making them, but need to check. If done properly, class a/b is really good. Conrad Johnson tube amps and krell amps do this. Amps and Preamps that use a combo of tubes and solid state have some real promise. CJ had a full class a pre that cost the first born and what was left of the wife.
I have been learning about guitar overdrive circuits and your videos have been much more useful than almost anything else I have read or viewed. Thank you!
_Why_? "They sound different because they are different." you say. That's awful close to a tautology. Not at all a why. A bunch of it is surely output transformers. Those things are non-linear as hell.
@Dave Micolichek It's called negative feedback network, and all amplifiers have it, it's something like that You: Amplified could you prepare me orange juice Amplifier: I could try, but I only have apples Negative feedback: I'm going to make you do orange juice Amplifier: Ok You: Thank you for the exquisite orange juice. I know it, every electronic engineer in the world knows it, even Paul knows it. Once you exceed a certain quality in the design, all the amplifiers sound the same, but keep the secret, we do not want to get into anyone's business.
@@joeking1019 I agree that at high signal levels that tubes clip differently than transistors, and this leads to differences in harmonic distortion. But in the video Paul claims that they sound different even when each is operating in their linear ranges. And in their linear ranges there should not be any harmonic distortion. Distortion only occurs in the non-linear ranges. Which leaves me wondering, who's correct, Paul or the people he claims are full of "you know what"? Perhaps it has more to do with the source impedance differences Paul mentions, and how those interact with the speaker? I'm not sure.
@@willdejong7763 Perhaps it is just a condition of being amplified through a vacuum tube, not necessarily driven to distortion, it was when I was younger and really into electronics and guitar amps that I learned about the 3rd harmonic produced in the tube, so I guess that just stuck as I've never heard a claim otherwise.
@giri kotte it could be both, this is from wikipedia Total harmonic distortion (THD) Music material contains distinct tones, and some kinds of distortion involve spurious tones at double or triple the frequencies of those tones. Such harmonically related distortion is called harmonic distortion. For high fidelity, this is usually expected to be < 1% for electronic devices; mechanical elements such as loudspeakers usually have inescapable higher levels. Low distortion is relatively easy to achieve in electronics with use of negative feedback, but the use of high levels of feedback in this manner has been the topic of much controversy among audiophiles.[citation needed] Essentially all loudspeakers produce more distortion than electronics, and 1-5% distortion is not unheard of at moderately loud listening levels. Human ears are less sensitive to distortion in the low frequencies, and levels are usually expected to be under 10% at loud playback. Distortion that creates only even-order harmonics for a sine wave input is sometimes considered less bothersome than odd-order distortion. Output power Output power for amplifiers is ideally measured and quoted as maximum Root Mean Square (RMS) power output per channel, at a specified distortion level at a particular load, which, by convention and government regulation, is considered the most meaningful measure of power
For me the biggest difference (as a guitar player) is the harmonics when distortion is present. The harmonics are frequencies that are added to the original signal. I don't know why but tubes produce even order harmonics which means that these new frequencies are higher octaves of the original signal. On the other hand transistors produce odd order harmonics which results in frequencies still dependent on the original signal but with in a different musical note.
Ed Berger solid state plus digital recording plus mp3 is like sticking an ice pick in your ears, and I don't have to know why it does to know that it does
I'm just about to restore an old HP oscilloscope. I pulled some of the tubes and listened to them... But nothing. My workshop is very quiet, but nothing. Not even a whisper, they laid there and said nothing. I think you should eat the honey instead of putting it into you ears. And yes, I do also prefer the sound of a tube amplifier. :-)
I believe that there are two major CAUSAL differences: (1)Tube amplifier distortion curves cross the cutting point of X and Y axis in a signal level/distortion graph. In other words, the lower the volume, lower the distortion. Thus at very low signal volume there is practically zero distortion. The transistor amplifier curve does not cross the meeting point of X and Y axis. The amplifiers never reach zero distortion. In some the distortion even increase at low volumes. Thus we always get some tangible distortion. (2) The noise levels of tube amplifiers are very low, specially at low volumes. The transistor always has some noise, so signal to noise ratio suffers at low signal levels. The audiophilic ears are extremely sensitive to distortion (and of course everyone can hear noise). That is why tube amplifiers sound so silky smooth and favored by audiophiles.
I commend Paul for not using the " transconductance curve " words.... He did a great job of describing whats going on without "loosing" the folks watching. BRAVO ! sbf
About 40 years ago an electronic technician I worked with who was also a musician, explained his simplistic rationale for the difference in sound. His take was that the heated filaments in the tubes would vibrate while being played. The variations in frequencies from the instruments imparted vibrations to the filaments of the tubes. This caused minute and constant changes in the flow of electrons between the various internal parts of the tubes. Solid state devices, having a different internal structure, lack the parts which enable the tubes to make the sound they do. 40 years ago, musicians were still talking about the sound that could only be found in a tube amplifier.
Its difficult to imagine resonance being a factor; if you compare the length of the valve elements with the wavelength of sounds at high frequency, @ 20Khz for example the wavelength is 17mm- maybe effects only the large valves?
I think you explain that quite well, from an electrical/ electronic point of view, harmonics play a big part in how many circuits work and how components interact, a slight variation in the right part of a circuit can make a huge difference to the output stage. A good and simple example of the harmonics effect can be found in computer networks... if you consider the plug into the wall powerline network adapters, they work by using high frequencies to send signals on cables already carrying electricity at lower frequencies (50 or 60 Hz), in modern telephone exchanges and fibre optic networks, this is done using MUX/ DMUX (multi plexer) systems where many frequencies can be sent as waves or pulses of light down the same cable/ fibre without interfering with each other. I have 2 Marshall guitar amps, a small solid state 15W practice amp that has a tube effect setting, it uses FETs and sounds pretty good and a 1978 SL100 series 1, it uses 4 tubes, 2 in the preamp and 2 larger ones in the power amp stage, I always joke that it sounds warmer, because the tubes have "heaters" used to excite the electrons, fact is, it does sound warmer, it has a natural distortion, even when playing "clean". Mr Carlson's Lab (UA-cam channel) has a good video on how tubes work).
T B, they where almost certainly talking about power amplifiers. Especially amps made for electric guitars where you deliberately drive the tube amp way out of the linear spectrum. You get tons of distortion from the amp, but its the kind of distortion they wanted. That is really far from what you want in a HiFi system where the idea is to reproduce the music as it is on the recording. Any kind of distortion your system adds will give a "sameness" on all music reproduced. With that said I should also say that a great clean producing HiFi system also lets makes all the badly recorded material very evident. It can often be to the point where you will avoid listening on badly produced material. Does that take the joy out of listening? I would say both yes and no. The best recordings will sound amazing and you will really enjoy them more. The bad recordings tend to be hard to listen to. If you still want to listen to them (and I do) the easiest way is to have another much cheaper system where you listen to them, e.g. i the car.
I love tube amplifiers and equipment. I still have my pair of Fisher 30 watt 80az amplifiers I found in the 1990s, and 400c preamp. All tube, from the late 1950s and fun to learn on and to service. I love listening to all my favorite music on it. Anything my heat desires. It’s been one of the most pleasurable discoveries in life for me. One thing you’ll notice, as you grow older, your hearing changes and you’ll start missing some of the effects and subtleties that make hi fi so fun. My ears in my late 40s aren’t what they use to be. Is why young ears are so wonderful. Anyway I love my tube stereo. Cheers!
Depending on the tube preamp and associated equipment in my experience tube front end mated to a solid state power amp produces a huge three dimensional sound stage with unmatched harmonics, especially in the mid range, that blow away solid state preamps. Voices, string instruments, reed instruments and pianos sound, to my ear, much, MUCH better than solid state gear. The price paid is loss of transient response, loss of bandwidth and a higher noise floor. But once ya hear tubes it's hard to go back to solid state.
Build something from the late lamented Allen Wright's "Tube Preamp Cookbook". No shortage of transient response or bandwidth there (although he does use FETs too).
@@vincentl.9469 If we're talking line stage preamps, a 4 tube amp will dissipate about 30 watts maximum. Is that too much heat? It isn't for me but your mileage may vary.
As an amature, the problem is that whenever I get to hear tubes - it's always on vinyl and horn loaded full range speakers. And when I listen the polar opposite it's: active speakers, DSP crossovers, class D playing Qobuz bit perfect FLAC 192/24 with full Dirac Live (like the Dynaudio Focus 50). So I think my idea of a "tube sound" is very much colored by the rest of the setup usually used. Add to that, the tube stack is playing some austere Kari Bremnes and the "tech stack" is playing Dream Theater or Tool.
Pierre Marcoux is a legendary audiophile in Quebec. He's been testing audio equipment for local manufacturers like Oracle. Here's what he told me : Tubes have distorsion on pair harmonics and transistors, on impair harmonics... Even though tubes have much higher distorsion level, it's not relevant because pair harmonics distorsion is not audible because of phase cancellation. In other words, the distorsion cancels itself, which is not the case with transistors. Transistors distorts on impair harmonics, which are audible. That's why transistors require feedback sent in the signal. While this lowers distorsion, it also removes some musicality. There were two other differencs, but I forgot what they we're... sorry!
Good explanation. I might add that even the "linear portion" of most transistor curves is somewhat "bowed", whereas there are tubes which (at least between cutoff and saturation) are almost ruler flat. Also, because tubes are such high impedance devices - (a fact you alluded to when you said they "don't draw current") - they present a very easy load to the input signal. The last thing I'd like to point out is that a transistor's characteristics vary GREATLY with temperature - (moreover, their capacitance can vary with input level) - whereas tubes are pretty consistent.
Good thing there are guys still around who dig messing with all of this stuff. Personally, I'll take most anything decent and try to make it sound good myself, but I can understand the hotrod aspect of playing with this stuff
The bottom line for the difference in tonal characteristics between valves and transistors is that valves tend to distort with a preponderance of even harmonics, and transistors tend to distort with a preponderance of odd harmonics. The even harmonic distortion is more musical because it is consonant with the original sound, whereas the odd harmonics are dissonant with the original sound. If you are always driving valves or transistors as a class A device and they remain _entirely_ within their linear range there should be no audible distinction between them because there should be no distortion of any sort. The differences occur when you drive them outside of their range of linearity, especially if you drive them either to the point of saturation or to turning off.
I did not study any of those thing. Thanks for make me to understand a so complicate system as a tube . I’m 52 years old ,and my childhood , I remember that we have to change the tubes in different electricdomestic artifacts.
I have heard that transistors tend to emphasize odd order harmonics, while tubes emphasize even order harmonics. Is this true? If so, why is that the case? Thanks for your great videos by the way, I'm learning a lot!
I started with tubes and when transistors came along, the major diference was the power curve and noise floor. Tubes have a much softer gain curve, and can amplify well past their max power limit and liniarity to the point of melt down. The resilting odd harmonics are easyer on the ear. Trensistors' sharper saturation point gives even harmonics (square waves) harsh on the ear when over driven. The design solution was to operate transistors well below saturation and prebias above the noise floor. Field effect transistors do act more like tubes, and as a result designers can now "program" amplifiers to sound like a tube amp.
The odd harmonics you describe are from push-pull circuits which cancel out even harmonic distortions and apply equally to solid state and tube amps. (Perhaps you meant EVEN harmonics?) The even harmonic distortions from a single tube power amp closely match the human ear. The brain interprets this distortion pattern as being louder, hence these tube amps sound louder than they measure. The sharp clipping distortion patterns from solid state gear is mostly down to the comparatively large amount of global negative feedback used. Pentodes have similar problems (plenty of gain at the expense of linearity). With simple speakers and triodes you can get away with no negative feedback at all. You get a natural sounding amp with dreadful measurements...
He said one is a voltage amplifying device and the other is a current amplifying device. It's the nature of how they operate that causes them to sound different.
ThePapabear2012 : Transistors aren't ever actually current amplifiers, that was just a scheme they came up with to escape a patent on FETs. The difference in sound comes down to differing harmonics, filtering, and biasing: if you bias a transistor or a tube correctly, then apply input, output, and feedback filtering correctly, then you CAN have them sound the same. The trick is that people don't go to sufficient effort to achieve the result (effectively, you're building a specialized analog computer for the purpose of simulating the "target sound": possible, but not necessarily worth the trouble).
ThePapabear2012 Jared Maddox uhhh... what? Amplifier manufacturers have been trying to make solid state amps sound like tube amps for decades (guitar amps, for example) and yet still there is a consensus among guitar players that tubes have a much richer, and therefore different sound. Tubes and transistors have very different physics, and very different sound characteristics. The current flowing in a tube does so in a near-vacuum, where the current in a FET flows through doped silicon. Tube current is controlled by a grid voltage, which produces an electrostatic force on the electrons which acts mostly parallel to the path of the electrons, whereas the FET modulates the charge between its source and drain terminals via a voltage applied at the gate which produces an electric field laterally to the path of the electrons, forming the channel between drain and source. As the poster mentioned, tubes operate at much higher voltages and temperatures than transistors as well, and of course the differences go on and on. They are two different ways to achieve the same goal but they do so differently, this is why they sound different.
William Ashland : And those characteristics (except for voltage, which can be handled if you're willing to spend the money) only matter because of their effect on the input characteristics (what are the input impedance characteristics, in essence), and output signal, which means that they can be simulated, including by analog semiconductor circuitry. The stuff you mentioned is all relevant to the designer of a device, but once a signal has passed through that device the effect no longer matters.
Amplifiers (and all audio devices) that actually sound different will most definitely measure different somehow. Anyone who says there are audible but inmeasurable differences is either full of you-know-what or in the business of manufacturing or selling those devices.
I agree FINALLYQQQQAVAILABLE. If an amplifying device is operating in its linear region then - by definition - its output is a linear amplification of its input, regardless of whether the device is a tube or transistor. Tube amps sound different to transistor amps because their circuits are different; fewer gain stages, capacitance or transformer interstages, less NFB and so on.
Nick T: You can take the output of any amplifying device and put it on an oscilloscope and you can see the waveform. If you have two outputs from two different devices and their waveforms look exactly the same, then those two waveforms *are* the same. - that's kinda the whole point of an oscilloscope: to compare waveforms.
Yes, generally we use a sine wave at blank hertz. A sine wave comparison at one frequency is not the same as music playback as the circuits an their parts change in linearity with different frequencies. I suggest you read nelson pass short write up on feedback, crossover distortion, and transistors. Also, look into the Hirata test.
Love your videos, my personal experence is they are all awesome, and both sound good in their own ways, and good for different types of music. I personally love tubes, from their slow warm up times to just simply how they work, to the sound they produce etc. Keep up the good videos :)
Everybody talks about frequency response when evaluating components or, in this case, tubes vs transistors. Linearity is related, for sure, as a nonlinear device will introduce harmonics. It’s easy to understand and the test is (relatively) easy to conduct. Input a sine wave and measure the output. But music isn’t a simple collection of continuous sine waves (keyword “continuous”). It is dynamic. From a drum beat to the pluck of a guitar string, every note is dynamic, that is, changing over time. Ever watch the vue meters on an amp? The ability of the gain stage (tube or transistor) to track these changes without distorting or running out of gas is critical. Tubes are voltage gain devices, almost no current draw on the input. Since no current flows, I suspect the tube is faster and better able to track the dynamics of the signal.
Great rundown on the differences. The generality is mostly correct. Enhancement Mode Field Effect Transistors do have very low input current in the DC area and can be operated with electrostatic charges. The difference here between them and vacuum tubes is a grid in a vacuum tube has very low capacitance as it is thin wire physically spaced away from everything else. In a Enhancement Mode FET, the gate capacitance is quite high, and even worse changes with voltage as the depletion region changes distance creating a distortion that tubes don't have. For this reason, FETs have to run with relatively high amounts of negative feedback and do not sound the same as tubes in low or no feedback applications. MOSFETS are best used as either switching devices in inverter power supplies such as automobile high power amps, or as output transistors with high levels of feedback to control linearity. Bipolar Junction Transistors, have various amounts of linearity depending on the device. Some are very linear and make great output stages in a classic AB biased push pull output stage with little feedback for linearity. The family of curves on these transistors can be seen when driven on a vintage Tektronix curve tracer. Some have excellent linearity over a family of current steps. www.valuetronics.com/product/370a-tektronix-curve-tracer-used?gclid=EAIaIQobChMIgsi3497s3AIVj19-Ch2jpQLQEAQYAiABEgKBFPD_BwE Our shop has one. Unfortunately the voltage step range is often exceeded by more modern MOSFETS with higher gate voltages. It would be nice if they updated the instrument to include modern MOSFETS. Some MOSFETS require more than 3 volts gate to begin conducting.
Nice explanation Paul. Anyway, yes they sound differently. But the main reason are the output transformers with their hysteresises and the slew rate. Due to the high value of the tube setup, they sound differently and gives you the warm feeling only tubes can give you. Signalwise on the paper go with BJT or with FETs if you like.
I researched Tubes and found that a scientific comparison had been done in a Lab. The results were most subjects prefered the music that was amplified by tubes. They said it was because of the fact that tubes have inherently a more natural distortion and that is more pleasing to our ears. So the report stated simply: The human ear preferes the distortion and final tones that tubes produce.
Interesting you mention that idea. I remember having a colour TV, bought way back in 1971, that had some tubes in it. It was in a small room and when it was on and the door closed it did a nice job of warming the room in the colder months of the year. I remember we would say our show is coming on soon, time to start up the heater!
Yep FET's and JFET's are very close to Tube smoothness ! That's why use them in my guitar overdrives. Transistors are Faster, that makes them great in front of a tube guitar amp.
The smaller Soviet tubes have very good lifespans (5000+ power on hours), so the "short lifespan" argument is not that distinct now (as well as these tubes being very readily available). There are a pretty large number of 6H3N and similar tubes appearing in the very modern range of hybrid amplifiers, and because the power dissipation of the "pre-amp" stage tubes is low, their lifespan is obviously far greater than that of power tubes (e.g. B300, KT88, etc).
yes tubes wear out but no one says you have to buy a tube amp with a large number of expensive tubes. i use a musical paradise set tube amp that takes all types and finding 2 new output tubes every couple years is not pricey and i find it fun. plus i can alter the sound 'quality' by using different types.
I bet much of it just comes down to size, as it takes longer for electrons to travel the length of the tube compared to the tiny distance of the silicon, you get different behaviour, for example, to electrons already in transit, when you charge the grid... so it's like the difference between a pneumatic system and a hydraulic system, where the pneumatic system softens changes as the gas can compress, whereas the incompressible hydraulic system responds very suddenly
Tubes sound different but they also measure different. You can of course claim something to sound different without measuring different just by not measuring what makes the difference. Tube amps are usually clipping the peaks softly and you get more musical distortion (even harmonics) than some hard clipping transistor amp. A simple FFT measurement or square wave comparison can easily prove this difference.
In the guitar amp world. There are a lot of units that use a tube preamp, and are solid state for the power section. A good 12ax7 triode-balanced tube will last a long time. A good tube like a Tung-sol etc, really adds to the preamp section in that saturation is achieved if pushed, as opposed digital clipping . That’s my 2 cents for the day ;)
4 роки тому
Each sounds good in its own way. Manufacturers of SS and Tube gear have successfully bridged the gap in differences in the quality of sound. The only real difference left between the two types of designs is that tube gear lacks the juice SS does when cranked up which is a serious issue for me.
Hi, I only understood that they are different. But, what is difference in sound between tubes and transistors, are tubes better at providing low frequency response?
Tubes are vacuum so they won't generate noise and reproduce true input signal due to very high input impedance (if you don't overdrive it), the only drawback is output transformer that dictates the frequency responds.
Sir Thank you for your knowledge and patience to try to explain the the very complex theory of electronics and sound You are in my opinion the best of what UA-cam can offer Thank you for what you do I’m truly a real fan
Good video explaining how tube amps work in the general sense. Though, I was distracted by the chirp of the smoke detector every 30 seconds or so. I kept counting the seconds till the next chirp.
There will always be tube amps that are made to sound “tubey” and transistor amps that sound “transistory”. From what I can hear though, the best of tubes and transistors are sounding more alike. As they get closer to the ideal they are converging toward each other as well. I do think the ultimate synergy will probably be the tube input-transistor output approach. Each device doing what it does best.
The real difference between tubes and MOSFETs are the input capacitance. And the internal impedance due to the load. And not to forget - tube amplifiers needs an output transformer to match (e.g.) an 8 ohm speaker, with the limitations (or advantages) hence to this fact.
I'm with you! Years of thinking, no preamp is better than a preamp, i bought a pass b1 (and it don't amplifies) and power amps got second ranging. Using a (in Germany Dynavox ET 100) Chip Class A/B Amp, it's as smooth and musically as the Musical Fidelity A1 Integrated, without the Class A power consumption. Also the dynamic is improved remarkable. Thanks for your Clip, i enjoyed it .
What about the difference in distortion? That is the primary reason WHY they sound different. Tubes produce even harmonics whereas transistors produce odd harmonics. Also, transistors have hard clipping whereas tubes have soft clipping. Their frequency response curves are also different. That is what gives tubes their "warm" sound.
Operating voltage range doesn't sound like a major reason for the sound differences. Tube amplifiers sound different because they have primarily second harmonic distortion (less harsh than transistor distortion), high output impedances (less damping on speaker resonances), commonly use single ended outputs, typically operate in class A mode, use transformers in the output stage and have modest feedback (if any). Reports from people who have better ears than I do suggest that tube amplifiers are hard put to compete with a good class A transistor amplifier - or even a basic one like the J Linsley Hood amplifier.
Obviously they sound "different" but a better question is: When the goal is purely linear amplification of an input signal (not intentional distortion like a guitar amp), which does so with the most faithful recreation of the input signal's characteristics?
Think less is more. What worked in the past works today. I am a converted tech head and realized the error of my ways. There is nothing like a good LP through a tube and some great speakers. It's a ritual , it pays homage to many decades of refining the design and it sounds natural and musical. There were engineering Giants in sound in those days. Newer seldom means better in the world of hi-fi.
My thoughts are that the human brain has the ability to determine if the input its getting is real or a facsimile. I think this is the core of the tube or transistor, analog or digital debate.Analog audio and tubes are a closer reproduction of the original ,even with its technological flaws.
Tubes and Bipolar Transistors sound different because they operate by completely different principles, the word "Transistor" is short for the devices full name "Transfer-Resistor".
@@suzesiviter6083 Yeah true, but tubes are mainly voltage-amplifier devices whereas solid-state transistors are mostly current-amplifier devices, yes there are Fets (Field Effect Transistors) and MOSFETS (Metal Oxide Semiconductor Field Effect Transistors) that operate very similar to the way a tube operates.
I’m sorry everybody but, when my 45 watt Solid state guitar amp was in the shop 40 years ago, I borrowed a buddies 6 watt Fender champ. That 6 watts of tube tone blew my 45 watt SS watt amp out of the water! I ditched that thing, used tube amps ever since and never looked back. So your Line 6 “models” a Tweed Bassman? Evidently, you’ve never played a Tweed Bassman!!!
Jack Dixon: I dont doubt it, but you know instruments like the guitar can sound better with added distortion, so this doesnt really apply to Hifi where you want the sound to be transparent.
@@suzesiviter6083 All good tube amps are super transparent, no audible distortion, what they actually call ''distortion'' is the non fatigue, sweet sound of tubes! Actually music sounds more real and live on tubes, especially jazz! But each to their own...
@@HASHEAVEN A good DSO will prove you wrong, Valves simply cannot go into the low THD figures a good bipolar or MOSFET amplifier produce; you prefer the sound and that's fine; but it is false to say you cannot hear the distortion relative to a good transistor design. Valves produce different types of harmonic distortion; this is one explanation why some people prefer Valves; IMO the warming up time and decreased life of tubes negates any small sonic advantage-but each to their own.
@@suzesiviter6083 I also like certain solid states like the simaudio moon, McIntosh, Accuphase and the clarity and detail of passlabs but those are all out of my reach. Maybe your are right but I certainly can't hear any distortion with my line magnetic, which I think sounds clearer and more detailed than many SS at it's price. Maybe what people call tube distortion is just the characteristic tube sound that I love... For ex I don't like the sound of an old SS Rotel receiver that I have, or a technics mosfet and a marantz of a friend and my LM 219 has better soundstage, detail bass midrange and more information on the highs than those, which for many people are good Solid state Amps.
@@HASHEAVEN Well whatever works for you, Valves do look sexy. I still have the old Pioneer A400 which was raved about in the 1980's; I haven't wished for an upgrade yet. Some of those old Amps can still keep up or better todays amps, the A400 for instance used custom silicon on the output transistors-which is rare on all but the most expensive amps even today. My biggest mistake was getting rid of my old Linn LP12; that had a certain magic and yes you can argue Vinyl has higher distortion; but the sound was just so much more enjoyable than a CD; so I know what you mean.
Valves are voltage-controlled devices. Normal transistors are current-controlled devices. FET transistors (Field Effect Transistor) are controlled in the same way as valves - by voltage.
thanks sir for all of these videos, there is alot of myths around this stuffs and what is better than that to see someone with practical experience explain it, every time you upload a video i can see i learned alot from it.
In transistors & all solid-state devices, the conduction of electricity is by "holes" & "non-holes" (electrons) which produce a hell lot of noise. But in vacuum tube, it is thro' pure electrons.
simple, tubes produces even/odd harmonics adding to the signal bec. it operates to be heated to function (filament/heater connections - 6.3v, etc.), whereas transistor/solid state device just needs some biasing in a fraction of a current in milliamps or millivolts. tubes plates voltages requires very high voltages (100v, 200v - or >1kv) defending on kinds of output tubes used in an amp and it's topology or design, i.e. single-ended or push-pull, class A, etc.....................
Being a tube guitar amp player of many, many years, it seems to me that tubes in the power section of an amp is what gives it its "balls". I understand, to some point, how tubes have a "softer,rounder" clip in the pre amp stage, but these hybrid amps that have a tubes(s) in the pre amp, and ss power, sound just as bad as all ss. Put a ss preamp in front of a full tube power amp, and you are mostly there. Of course, all tube is best, but man i am getting too old to carry around a 95 pound 2x12 combo. I would love to find a ss amp that actually delivers.
Not really a valid argument. most guitar amplifiers are designed to colour the sound. They are a part of the overall instrument. Seems to be some confusion as to warm reproduction and accurate reproduction Not possible to have both as the warmth is distortion. it might sound good but it aint HIFI
Perhaps I missed something. I'm 66 years old, have owned nice hi-fi equipment most of my life. I didn't get how they sound different from one another? What exactly is it that makes them sound different? I know tubes are supposed to have a "warmer sound," maybe "smoother" or "rounder" are better terms. What are we hearing precisely?
I always said the control signal in tubes is higher than the audio signal. In solid state the control is lower than the audio signal. So when you pump up the volume on ss devices, the audio signal seeps into the control, then you get bad clipping.
Is there a significant difference between using a 12 meter 16 gauge silverplated audio cables vs 11 gauge ofc copper cable that is short? despite the resistance which difference is 0.3 ohms which probably isnt even noticable and wont cause any signal loss maybe a small percentage of power loss right?
I think you said even if the data shows the plots the same they still sound different and if so I find it hard to believe. I am curious if you ever did a blindfold test lol and how you did? Would be interesting. I use mosfets to pwm control motors so was wondering if you do that with speakers? If so you could linearize if needed or closed loop servo as well which is what we do. Or you could make a lookup table to replicate the tube response. Would love to know. Thanks for the video!
Howdy. One possible explanation. In a BJT the collector current is dependent only of the base current and the gain factor. Collector voltage has no effect on the collector current. Within usable limits, of course. Tubes are different. The anode voltage will have some effect on the anode current. The gain factor is not a constant. The gain factor will be smaller at high signal when the anode resistor causes the anode voltage to be lower. This is an inherent compressing property of tubes. It introduces the third harmonic to the signal. Regards.
@@marianneoelund2940 Hi. Blimey. I stand corrected. The collector current is truly slightly dependent on the collector voltage. Thanks for the correction. In high regards.
8:25 "anybody who will tell you that [three devices] ... running at the same [specified parameters] ... do not sound different because they don't _measure_ different ... is [wrong]" You never did explain exactly how or why in detail. I'm still not sold on the argument that tubes > transistors. But my focus is on the "measurement" itself. If instruments have finer precision and better resolution than the human ear ... and they "measure the same" ... then how can the things they measure "sound" different to the human ear?
The thing is that most music producers will make make it so the sound that we ear has tube sound being played by solid state amps... which renders all this discussions immaterial.
Paul's explanation is somewhat incorrect. Tubes only function in a vacuum because air blocks the electron flow between cathode and anode/plate (beyond the filament burning up). To my understanding, filaments usually are tungsten or an alloy of tungsten and other elements, and the cathode (the part that actually does the thermionic emission/"boiling off of electrons") is made of a combination of alkaline earth metal oxides (e.g. barium oxide or strontium oxide).
Tubes do sound differently. I have a 1957 The Fisher model 80T Hi-Fi mono AM/FM tuner and preamp. In addition to treble and bass control it has four loudness levels. I connected it to a solid state integrated amplifier. Whenever I want to listen to old hi-fi jazz or blues recordings, I use the 80T. The sound is far richer and warmer than that of the solid state own preamp, although it is mono. The technicians of The Fisher went to great lengths to have a hi-fi preamp with very low noise floor and great Dynamic Range. The four levels of loudness allied with the treble and bass controls provide the best audio configuration. This 80T preamp and tuner also came with a power amplifier which also has it's own four level loudness control. I haven't used yet, because I have to figure out a way of commuting the speaker boxes from my solid state strereo equipment to this mono power amplifier and vice-versa.
Great Videos Paul! You have cleared up some understanding of gear and how things work! Thank You for making these videos. If i get out to CO. I will be paying yall a visit!
Start again from basics, please. A bjt is a linear current amplifier (current in-current out, small input impedance, large output impedance), a triode is a linear voltage amplifier (voltage in-voltage out, large input impedance, low output impedance), a jfet/mosfet/pentode is a linear transconductance amplifier (voltage in-current out, large input impedance, large output impedance). A loudspeaker driver is a motor, therefore driven by current (to behave linearly). But then, there is the loudspeaker crossover circuit, which is usually designed as a voltage devider, unfortunately. From these basics, add the high voltage bias of tubes (that increases electric carriers mobility) and tubes' "excess" amount of carriers (as electronic cloud inside them) that helps in soft clipping at large transient demands. From the above all combined, go to amplifier's output driver stage, then to the input stage and finally combine them all together, keeping in mind that we are used to design (from the beginning) voltage mode circuits that are easier (in controlling voltage). We are also used to voltage signals, but that's another long story. Have a nice day, all.
Not quite following you Yiorgis Alexakis. By triode do you mean vacuum tube or valve triode? Because tubes have high output impedance, which is why they typically use output transformers to connect to loudspeakers.
Stuve715, you refer to "high impedance" in relation (relative) to the loudspeaker impedance, which is quite lower. Driving a relatively lower impedance, is what makes necessary the use of output transformer (and that does not contradict to the "low output impedance" concept of a voltage amplifier). In general, a triode tube/valve- being a voltage amplifier - should present a "low" output impedance (if it is a good voltage amplifier) to prevent action of a voltage divider between its output impedance and the load impedance (something that alters output voltage on the load, thus "violating" the constant voltage source concept of a voltage amplifier's output. I started my comment by saying that going back to basics could explain a lot of vague impressions that survive in audio in general. For example, all tubes do not have the same amplifying behavior. Pentodes are not voltage amplifiers, as triodes. They are transconductance amplifiers. Another one: higher voltage JFETs/MOSFETs are quite different from low voltage ones as far as transconductance linearity and dynamic range (but that is another long story).
I wish you did the video, and explain current amplifiers as I always think volts is the energy difference that causes the current, can't think of a flowing current causing a voltage. Also, I thought you want high impedance i/p so the volts at i/p not affected too much ? Also isn't it something to do with harmonics why valve clipping is better ?
Starting from your last question, harmonics is a phenomenological result (that we measure) of what is actually happening inside the tube and that is the excess of electrons stored as a cloud and the current demands they can fulfill. These electrons contribute to soft clipping and also when a tube is over driven. We refer to the larger percentage of even-to-odd harmonics of the tube's output to just distinguish the perception that even harmonics (governing tube's output) give "light" to a more acceptable sound to our ears (odd harmonics are sounding harsh to our ears). Harmonics are produced from the non-linearity of an amplifier's output current-to-voltage characteristic (abrupt non-linearity produces more odd harmonics), which also has to do with the availability of carriers demanded by the signal to be driven to the amplifier's output load. Now, as far as high input impedance, it is indeed necessary for avoiding voltage signal losses at the amplifier's input. It is a "must" when our input signals are voltage signals, especially small ones. Higher input impedances can be found in amplifying devices driven by voltage, such as voltage amplifiers (e.g. triodes, opamps) and transconductance amplifiers (e.g. jfets, mosfets, pentodes). As for the current amplifiers, they need to be driven by current at their input as their input signal, in order to behave as linear amplifiers at their output. The device example we have is the BJT. Finally, as you mention, voltage is the potential difference that is needed for the current to flow in between. But "free" moving carriers are needed in the first place for the current to be created. This is general and it doesn't contradict with the case of current amplifiers, at all. I hope I did give you at least a rough idea about the subject. The limited space here is not suitable for more details. You can find many resources about all types of amplifiers at the internet to continue your quest.
Thank you very much for answering my questions, I can see that you have a great deal of insight in this subject, and I will look into BJT current amplifiers more now, thank you.
Thank you for sharing, very interesting and informative, just uploaded a few of my hand crafted tube Amps. I enjoy learning more about the world of tube Audio!
A tube amp and transistor amp should sound identical if well designed, equal output, and operated within their linear specifications. If you start to drive them out of non-linearity, all bets are off.
Tubes DO typically sound different... because they introduce additional nonlinearities into the system beyond what a FET or BJT amplifier would. Tubes warm the sound by distorting even order harmonics and allow for "soft clipping" by a similar mechanism. You can just about digitally emulate the "tube sound" at this point, but it's still probably worth it for studios to record with actual tubes. However, for playback I strongly prefer a transparent sound... which typically means solid state.
The tubes cushion the sound, that is, they dampen it. I don't think it's any simpler than that. Why do they sound better? I'm not an engineer or an electronics technician, but starting from the fact that when an electrical signal reaches the tube it has to be converted into electrons and pass through a gaseous medium, perhaps it is this medium that adds dynamics to the flow of electrons.
One interesting thing I have noticed empirical. If you replace one BJT with another in an amplifier, the sound will be more or less the same even if the transistor is of another brand and have other specs( within reasonable limits ). But if you replace a tube with another one, the sonic imprint usually is totally changed. That is what I call interesting. And a bit of a mystery.
vacume tubes surley will be sounding different every second they are played . Or am i wrong and they wear out in stages. Come to think of it it dosnt even need to be used as some deterioration is happening due to contaminents that remain in a non perfect vacume??
I was a Hi-Fi snob. But some time ago I realized I was listening more and enjoying it less. I envied people who had what I considered crap systems, but who got great enjoyment out of listening. I had my moment of truth; I was chasing my tail, I wasn't listening to MUSIC anymore, I was listening to THE EQUIPMENT! That's two different things. The whole point is the enjoyment of music. It's one thing if your equipment is defective, but does one need to spend$5,000, $10,000, $20,000 or more before they can start enjoying music? Don't get me wrong, of course I would love to have that $50,000+ system with a room to match. I'm just saying, don't get lost, and don't miss the point of all this, which is THE ENJOYMENT OF MUSIC.
Well said, it's called the law of diminishing returns.
MrMinathecat You are so right. In fact I have been an audiophile for over 35 years and have always noticed that most audiophiles listen to music that sounds good on their systems and not necessarily the music they like best.
I also used to play with audiophile systems but not anymore. It's good to listen to that excellent systems but I found myself enjoying various music not recorded up to audiophile or even HiFi standard. I just enjoy more music than the quality of the reproduced sound. If it's good music, you can feel it's good music, no matter how good it was recorded and reproduced out of the speakers. The players have put their souls in the piece already and the audience can absorb the messages musicians wanted to convey. Most musicians I know do not care much about their listening systems. They usually use whatever they can find to enjoy music and they can still distinguish the good, the mediocre, and the bad pieces meticulously.
I still have my hi-end system but I'm not playing around with it anymore. Just consider it as an excellent system to play recorded music sometimes I want to enjoy a better quality.
a couple of grand worth spent should get you a good sound system.
Personally, (just my opinion) a $5,000 - $10,000 hifi system will give you some very good musical experience, regardless of genre.
Spending more will give you some improvement, but not much.
I've been waiting 25 years for that, lol. I really appreciate it. That was the simplest, most concise explanation of valves vs solid state I've heard.
To the guy who dosent understand how it effects sound: remember that voltage = PSI (roughly) running at higher voltages creates a higher share of linearity for the signal. The more linearity, the more headroom. More headroom allows for broader soundstage (x&y), faster reactions to peaks and lows, it adds some warmth (although a good SS amp that's good and warm and running in the meat of it's powerband is hard to beat)
and definitely "puts less strain" on the signal b/c of the exponentiated section of linearity on offer to the signal.
Thanks for this kind comment. Glad I could help.
Thanks for posting this Paul. Much appreciated. Both you and I are men of a certain age. I am a vacuum-tube enthusiast. I have a few vintage-1962 Dynaco ST-70's and a few Dynaco PAS-3 preamps that I've kept alive for 40+ years. Thanks again for your expert explanation of tubes vs. transistors.
TRUTH to tell, they LOOK cool. Something nostalgic about that warm glow makes me feel good.
I used vacuum tube Zenith Cobramatic all thru college. Acid Rock like Cream,Jethro Tull,etc sounds like 1968
"Something nostalgic about that warm glow makes me feel good."
===
I was packed off to boarding school in the late 1960s and took my Dad's cast-off tube audio gear with me. A number of my music-loving classmates had similar stuff. All our parents had the latest shiny transistor equipment but it didn't sound so good to us.
Over the years I have had some really nice solid state gear but 25 years ago I built my first single triode power amp. It felt like "coming home" and is still my first choice, despite the more costly solid state gear I have lying around. Built to satisfy rather than impress...
@@TheRealWindlePoons Does this satisfaction lies in the ritual, the tradition or anything else that has more to do with some social aspect than the music itself?
@@mignotmaxime2409 No. Not Ritual. Not tradition. Not social.
The single triode amplification stage has a natural presentation I have seldom found elsewhere. But its ALWAYS about the music. Otherwise, what's the point?
@@TheRealWindlePoons Thanks. I just read the 10 biggest lies in HiFi by Peter Aczel who says tubes do not sound different. I was wondering if the preference for tubes could be elsewhere than the music. Not necessarly a bad thing but it is the case, it must be clearly stated.
Thanks for this. I just came across two young tube enthusiasts in my neighbourhood building their own preamps. I'm an RF/CE engineer and was heartened to see young people getting "hands-on" in this age of buy and throw away. I believe the sound difference is in the even octaves only that is found in tubes, whereas the transistors let thru both odd and even harmonics. There's a music theory that says even harmonics makes a pleasant sound, why some music tunes are more appealing than others.
Second, fourth, sixth and eighth harmonics are all musically harmonious. But with odd harmonics, only the third qualifies. The fifth, seventh, etc., do not form accurate musical intervals, so they cause dissonance. In addition, distortion from solid state designs tends to produce very high-order harmonics, causing a raspy, crackly or buzzing effect.
Interestingly enough the human ear actually produces even harmonic distortion itself.
Yes, they sound different, but "different" doesn't necessarily mean better. In fact the coloration added by Tubes (pleasant or not) is often referred to as "good distortion"
Tubes are used in the recording studio too. Some microphones use Tubes to warm the sound. And Jazz & Blues guitarist use Tubes for the sound they produce when over driven. I'm going to kick my own ass for saying this, but Tubes can make certain digital recordings sound better to our ears, BUT it greatly depends on what's being recorded.
Wrong ..............tubes are not warm sounding , only if not correctly implemented or wrongly used, clearly you don't know about vacuum -tube technology and that's a fact ............
Frank Geeraerts
We use Tube mics in the studio to to warm or smooth an instrument. The terminology is just geek speak as long as its understood during sessions. Vocals can sound very nice with Tube pre's ... The Hi-Fi world is rather different. I wouldn't use Tubes in my system as they're not needed.
And Frank, think about how you reply 😉
Chris Bishop its obviously a joke comment
Frank, tube amps MAY sound "warmer" if they're single-ended design, in which case you get a noticeable portion of the 2-nd harmonic (over -60dB), which does sound somewhat pleasant. You don't get that effect from a push-pull design.
Chris Bishop if
Two reasons tubes sound different, one is soft saturation, tubes can operate into a long saturation curve, two is that output is "almost" always through a transformer with hysteriesis distortion.
When newer fet amps are designed to sound like a tube amp these "distortions" are programmed into the design.
This whole thing cracks me up. I'm a guitar player. I've used my old Peavey guitar amp on gigs and had other guitar players come up to me and say "... man, that thing sounds great.... that must be a tube amplifier... " I like this because I'm using a relatively inexpensive amp, that has no tubes in it. In fact, the whole amp (which is all discrete, no ICs) has only one FET in it. The rest of that transistors are bipolar junction types (BJT transistors.) Peavey was cleaver in that they did use an audio transformer on the output, which helps to add just a bit of tube like "mud" into the sound. This amp was a cheap amp, to begin with (and I've replaced some parts with parts from a surplus store, occasionally) and that's what makes is especially amusing when guitar players mistake this for a tube amp. After I reveal that this is a solid state amp, I say to them: If tubes, alone, made for audiophile level listening, every table radio from the 30s, 40s and 50s would be audiophile quality... and we know that ain't true.
cool, I always liked the way old vacuum tube radios sounded when I was a kid. As an kid/teenager during the 1960's We guys use to hangout and go everywhere with our little 9 volt transistors. Happy to hear that the vacuum tube has not gone the way of the horse.
Yes, true said
Audio Research went back and forth between tubes and transistors in their amps and preamps and their tube designs were always better. Warmest sounding amp I have is a transistor amp made in Norway but it gets hot as a stove. Tubes don't really have a warm sound. But what they do have is a lack of that hard edge sound of alot of transistor amps have.Everything colors sound but tubes tend to sound more like the real thing.
Class A amplification always sounds better, even if it is solid state. JVC had some really good amps that were super cheap (less than $100) , but they had poor speakers. I suppose they quit making them, but need to check. If done properly, class a/b is really good. Conrad Johnson tube amps and krell amps do this. Amps and Preamps that use a combo of tubes and solid state have some real promise. CJ had a full class a pre that cost the first born and what was left of the wife.
I have been learning about guitar overdrive circuits and your videos have been much more useful than almost anything else I have read or viewed. Thank you!
_Why_?
"They sound different because they are different." you say. That's awful close to a tautology. Not at all a why.
A bunch of it is surely output transformers. Those things are non-linear as hell.
otl amp it is than
@@Redh0und High output impedance
@@johnyang799 depends
Could be why McIntosh use autoformers on their higher end solid state power amps.
@Dave Micolichek
It's called negative feedback network, and all amplifiers have it, it's something like that
You: Amplified could you prepare me orange juice
Amplifier: I could try, but I only have apples
Negative feedback: I'm going to make you do orange juice
Amplifier: Ok
You: Thank you for the exquisite orange juice.
I know it, every electronic engineer in the world knows it, even Paul knows it.
Once you exceed a certain quality in the design, all the amplifiers sound the same, but keep the secret, we do not want to get into anyone's business.
I think you should consider even vs. odd harmonics in tube vs. solid state systems.
Yes they are different, BUT you never got to the WHY they sound different.
3rd harmonic distortion
@@joeking1019 I agree that at high signal levels that tubes clip differently than transistors, and this leads to differences in harmonic distortion. But in the video Paul claims that they sound different even when each is operating in their linear ranges. And in their linear ranges there should not be any harmonic distortion. Distortion only occurs in the non-linear ranges. Which leaves me wondering, who's correct, Paul or the people he claims are full of "you know what"? Perhaps it has more to do with the source impedance differences Paul mentions, and how those interact with the speaker? I'm not sure.
@@willdejong7763 Perhaps it is just a condition of being amplified through a vacuum tube, not necessarily driven to distortion, it was when I was younger and really into electronics and guitar amps that I learned about the 3rd harmonic produced in the tube, so I guess that just stuck as I've never heard a claim otherwise.
@giri kotte it's not worth an argument if you offer nothing to back it up,
@giri kotte it could be both, this is from wikipedia
Total harmonic distortion (THD)
Music material contains distinct tones, and some kinds of distortion involve spurious tones at double or triple the frequencies of those tones. Such harmonically related distortion is called harmonic distortion. For high fidelity, this is usually expected to be < 1% for electronic devices; mechanical elements such as loudspeakers usually have inescapable higher levels. Low distortion is relatively easy to achieve in electronics with use of negative feedback, but the use of high levels of feedback in this manner has been the topic of much controversy among audiophiles.[citation needed] Essentially all loudspeakers produce more distortion than electronics, and 1-5% distortion is not unheard of at moderately loud listening levels. Human ears are less sensitive to distortion in the low frequencies, and levels are usually expected to be under 10% at loud playback. Distortion that creates only even-order harmonics for a sine wave input is sometimes considered less bothersome than odd-order distortion.
Output power
Output power for amplifiers is ideally measured and quoted as maximum Root Mean Square (RMS) power output per channel, at a specified distortion level at a particular load, which, by convention and government regulation, is considered the most meaningful measure of power
For me the biggest difference (as a guitar player) is the harmonics when distortion is present. The harmonics are frequencies that are added to the original signal. I don't know why but tubes produce even order harmonics which means that these new frequencies are higher octaves of the original signal. On the other hand transistors produce odd order harmonics which results in frequencies still dependent on the original signal but with in a different musical note.
Listening to tubes is like pouring honey in your ears.
That sounds like a terrible idea unless you want to be covered with ants.
Ed Berger solid state plus digital recording plus mp3 is like sticking an ice pick in your ears, and I don't have to know why it does to know that it does
I'm just about to restore an old HP oscilloscope. I pulled some of the tubes and listened to them... But nothing. My workshop is very quiet, but nothing. Not even a whisper, they laid there and said nothing. I think you should eat the honey instead of putting it into you ears.
And yes, I do also prefer the sound of a tube amplifier. :-)
I love honey!
Then you end up hearing nothing, which in case of nowadays music, is quite an improvement.
I believe that there are two major CAUSAL differences: (1)Tube amplifier distortion curves cross the cutting point of X and Y axis in a signal level/distortion graph. In other words, the lower the volume, lower the distortion. Thus at very low signal volume there is practically zero distortion. The transistor amplifier curve does not cross the meeting point of X and Y axis. The amplifiers never reach zero distortion. In some the distortion even increase at low volumes. Thus we always get some tangible distortion.
(2) The noise levels of tube amplifiers are very low, specially at low volumes. The transistor always has some noise, so signal to noise ratio suffers at low signal levels.
The audiophilic ears are extremely sensitive to distortion (and of course everyone can hear noise). That is why tube amplifiers sound so silky smooth and favored by audiophiles.
I commend Paul for not using the " transconductance curve " words.... He did a great job of describing whats going on without "loosing" the folks watching.
BRAVO !
sbf
Hmm...that actually gives insight
About 40 years ago an electronic technician I worked with who was also a musician, explained his simplistic rationale for the difference in sound. His take was that the heated filaments in the tubes would vibrate while being played. The variations in frequencies from the instruments imparted vibrations to the filaments of the tubes. This caused minute and constant changes in the flow of electrons between the various internal parts of the tubes. Solid state devices, having a different internal structure, lack the parts which enable the tubes to make the sound they do. 40 years ago, musicians were still talking about the sound that could only be found in a tube amplifier.
Its difficult to imagine resonance being a factor; if you compare the length of the valve elements with the wavelength of sounds at high frequency, @ 20Khz for example the wavelength is 17mm- maybe effects only the large valves?
I think you explain that quite well, from an electrical/ electronic point of view, harmonics play a big part in how many circuits work and how components interact, a slight variation in the right part of a circuit can make a huge difference to the output stage. A good and simple example of the harmonics effect can be found in computer networks... if you consider the plug into the wall powerline network adapters, they work by using high frequencies to send signals on cables already carrying electricity at lower frequencies (50 or 60 Hz), in modern telephone exchanges and fibre optic networks, this is done using MUX/ DMUX (multi plexer) systems where many frequencies can be sent as waves or pulses of light down the same cable/ fibre without interfering with each other.
I have 2 Marshall guitar amps, a small solid state 15W practice amp that has a tube effect setting, it uses FETs and sounds pretty good and a 1978 SL100 series 1, it uses 4 tubes, 2 in the preamp and 2 larger ones in the power amp stage, I always joke that it sounds warmer, because the tubes have "heaters" used to excite the electrons, fact is, it does sound warmer, it has a natural distortion, even when playing "clean".
Mr Carlson's Lab (UA-cam channel) has a good video on how tubes work).
T B, they where almost certainly talking about power amplifiers. Especially amps made for electric guitars where you deliberately drive the tube amp way out of the linear spectrum. You get tons of distortion from the amp, but its the kind of distortion they wanted.
That is really far from what you want in a HiFi system where the idea is to reproduce the music as it is on the recording.
Any kind of distortion your system adds will give a "sameness" on all music reproduced.
With that said I should also say that a great clean producing HiFi system also lets makes all the badly recorded material very evident. It can often be to the point where you will avoid listening on badly produced material.
Does that take the joy out of listening?
I would say both yes and no. The best recordings will sound amazing and you will really enjoy them more. The bad recordings tend to be hard to listen to. If you still want to listen to them (and I do) the easiest way is to have another much cheaper system where you listen to them, e.g. i the car.
@@Rohan4711 Great post, especially your reference to the cheaper system.🙂
I love tube amplifiers and equipment. I still have my pair of Fisher 30 watt 80az amplifiers I found in the 1990s, and 400c preamp. All tube, from the late 1950s and fun to learn on and to service. I love listening to all my favorite music on it. Anything my heat desires. It’s been one of the most pleasurable discoveries in life for me. One thing you’ll notice, as you grow older, your hearing changes and you’ll start missing some of the effects and subtleties that make hi fi so fun. My ears in my late 40s aren’t what they use to be. Is why young ears are so wonderful. Anyway I love my tube stereo. Cheers!
Depending on the tube preamp and associated equipment in my experience tube front end mated to a solid state power amp produces a huge three dimensional sound stage with unmatched harmonics, especially in the mid range, that blow away solid state preamps. Voices, string instruments, reed instruments and pianos sound, to my ear, much, MUCH better than solid state gear. The price paid is loss of transient response, loss of bandwidth and a higher noise floor. But once ya hear tubes it's hard to go back to solid state.
Build something from the late lamented Allen Wright's "Tube Preamp Cookbook". No shortage of transient response or bandwidth there (although he does use FETs too).
@@TheRealWindlePoons the drawback is the amount of heat tubes produce...
@@vincentl.9469 If we're talking line stage preamps, a 4 tube amp will dissipate about 30 watts maximum. Is that too much heat? It isn't for me but your mileage may vary.
what I like about the tube, is the ambience it brings into the music. have it on DAC and pre amp, although still using solid state power amp.
As an amature, the problem is that whenever I get to hear tubes - it's always on vinyl and horn loaded full range speakers.
And when I listen the polar opposite it's: active speakers, DSP crossovers, class D playing Qobuz bit perfect FLAC 192/24 with full Dirac Live (like the Dynaudio Focus 50).
So I think my idea of a "tube sound" is very much colored by the rest of the setup usually used.
Add to that, the tube stack is playing some austere Kari Bremnes and the "tech stack" is playing Dream Theater or Tool.
Pierre Marcoux is a legendary audiophile in Quebec. He's been testing audio equipment for local manufacturers like Oracle. Here's what he told me : Tubes have distorsion on pair harmonics and transistors, on impair harmonics... Even though tubes have much higher distorsion level, it's not relevant because pair harmonics distorsion is not audible because of phase cancellation. In other words, the distorsion cancels itself, which is not the case with transistors. Transistors distorts on impair harmonics, which are audible. That's why transistors require feedback sent in the signal. While this lowers distorsion, it also removes some musicality. There were two other differencs, but I forgot what they we're... sorry!
Good explanation. I might add that even the "linear portion" of most transistor curves is somewhat "bowed", whereas there are tubes which (at least between cutoff and saturation) are almost ruler flat.
Also, because tubes are such high impedance devices - (a fact you alluded to when you said they "don't draw current") - they present a very easy load to the input signal.
The last thing I'd like to point out is that a transistor's characteristics vary GREATLY with temperature - (moreover, their capacitance can vary with input level) - whereas tubes are pretty consistent.
so we build an amplifier with a ruler flat curve . Then we add a BBC or some other dip to the loudspeakers .. Really!!
Ruler flat depends how large the paper and ruler is you are using!.
They sound different because they are different.
Thanks Paul, that really cleared things up.
Good thing there are guys still around who dig messing with all of this stuff. Personally, I'll take most anything decent and try to make it sound good myself, but I can understand the hotrod aspect of playing with this stuff
The bottom line for the difference in tonal characteristics between valves and transistors is that valves tend to distort with a preponderance of even harmonics, and transistors tend to distort with a preponderance of odd harmonics. The even harmonic distortion is more musical because it is consonant with the original sound, whereas the odd harmonics are dissonant with the original sound.
If you are always driving valves or transistors as a class A device and they remain _entirely_ within their linear range there should be no audible distinction between them because there should be no distortion of any sort.
The differences occur when you drive them outside of their range of linearity, especially if you drive them either to the point of saturation or to turning off.
The point of tubes is using them outside their range of linearity in order to create 'euphonic' low order distortion.
@@danaustin5869 That is one way to use valves. Many folk are more interested in eliminating distortion entirely.
I did not study any of those thing. Thanks for make me to understand a so complicate system as a tube . I’m 52 years old ,and my childhood , I remember that we have to change the tubes in different electricdomestic artifacts.
I have heard that transistors tend to emphasize odd order harmonics, while tubes emphasize even order harmonics. Is this true? If so, why is that the case? Thanks for your great videos by the way, I'm learning a lot!
I started with tubes and when transistors came along, the major diference was the power curve and noise floor. Tubes have a much softer gain curve, and can amplify well past their max power limit and liniarity to the point of melt down. The resilting odd harmonics are easyer on the ear.
Trensistors' sharper saturation point gives even harmonics (square waves) harsh on the ear when over driven.
The design solution was to operate transistors well below saturation and prebias above the noise floor.
Field effect transistors do act more like tubes, and as a result designers can now "program" amplifiers to sound like a tube amp.
The odd harmonics you describe are from push-pull circuits which cancel out even harmonic distortions and apply equally to solid state and tube amps. (Perhaps you meant EVEN harmonics?)
The even harmonic distortions from a single tube power amp closely match the human ear. The brain interprets this distortion pattern as being louder, hence these tube amps sound louder than they measure.
The sharp clipping distortion patterns from solid state gear is mostly down to the comparatively large amount of global negative feedback used. Pentodes have similar problems (plenty of gain at the expense of linearity). With simple speakers and triodes you can get away with no negative feedback at all. You get a natural sounding amp with dreadful measurements...
@@TheRealWindlePoons very true said, always misguided the issue by the uneducated people. Thanks brother
You certainly reinforced the fact THAT they do, but not so much WHY they do.
Richard Larson yes he did. Did you watch the video?
He said one is a voltage amplifying device and the other is a current amplifying device. It's the nature of how they operate that causes them to sound different.
ThePapabear2012 : Transistors aren't ever actually current amplifiers, that was just a scheme they came up with to escape a patent on FETs.
The difference in sound comes down to differing harmonics, filtering, and biasing: if you bias a transistor or a tube correctly, then apply input, output, and feedback filtering correctly, then you CAN have them sound the same. The trick is that people don't go to sufficient effort to achieve the result (effectively, you're building a specialized analog computer for the purpose of simulating the "target sound": possible, but not necessarily worth the trouble).
ThePapabear2012 Jared Maddox uhhh... what? Amplifier manufacturers have been trying to make solid state amps sound like tube amps for decades (guitar amps, for example) and yet still there is a consensus among guitar players that tubes have a much richer, and therefore different sound.
Tubes and transistors have very different physics, and very different sound characteristics. The current flowing in a tube does so in a near-vacuum, where the current in a FET flows through doped silicon. Tube current is controlled by a grid voltage, which produces an electrostatic force on the electrons which acts mostly parallel to the path of the electrons, whereas the FET modulates the charge between its source and drain terminals via a voltage applied at the gate which produces an electric field laterally to the path of the electrons, forming the channel between drain and source. As the poster mentioned, tubes operate at much higher voltages and temperatures than transistors as well, and of course the differences go on and on.
They are two different ways to achieve the same goal but they do so differently, this is why they sound different.
William Ashland : And those characteristics (except for voltage, which can be handled if you're willing to spend the money) only matter because of their effect on the input characteristics (what are the input impedance characteristics, in essence), and output signal, which means that they can be simulated, including by analog semiconductor circuitry. The stuff you mentioned is all relevant to the designer of a device, but once a signal has passed through that device the effect no longer matters.
Amplifiers (and all audio devices) that actually sound different will most definitely measure different somehow. Anyone who says there are audible but inmeasurable differences is either full of you-know-what or in the business of manufacturing or selling those devices.
I agree FINALLYQQQQAVAILABLE. If an amplifying device is operating in its linear region then - by definition - its output is a linear amplification of its input, regardless of whether the device is a tube or transistor.
Tube amps sound different to transistor amps because their circuits are different; fewer gain stages, capacitance or transformer interstages, less NFB and so on.
Nick T: You can take the output of any amplifying device and put it on an oscilloscope and you can see the waveform. If you have two outputs from two different devices and their waveforms look exactly the same, then those two waveforms *are* the same. - that's kinda the whole point of an oscilloscope: to compare waveforms.
Yes, generally we use a sine wave at blank hertz. A sine wave comparison at one frequency is not the same as music playback as the circuits an their parts change in linearity with different frequencies. I suggest you read nelson pass short write up on feedback, crossover distortion, and transistors. Also, look into the Hirata test.
linearaudio.nl/sites/linearaudio.net/files/Hirata%20test%20engineering%20report%201981.pdf
Read this.
Nick T: At what point did I say sine wave?
Love your videos, my personal experence is they are all awesome, and both sound good in their own ways, and good for different types of music. I personally love tubes, from their slow warm up times to just simply how they work, to the sound they produce etc. Keep up the good videos :)
Everybody talks about frequency response when evaluating components or, in this case, tubes vs transistors. Linearity is related, for sure, as a nonlinear device will introduce harmonics. It’s easy to understand and the test is (relatively) easy to conduct. Input a sine wave and measure the output.
But music isn’t a simple collection of continuous sine waves (keyword “continuous”). It is dynamic. From a drum beat to the pluck of a guitar string, every note is dynamic, that is, changing over time. Ever watch the vue meters on an amp? The ability of the gain stage (tube or transistor) to track these changes without distorting or running out of gas is critical.
Tubes are voltage gain devices, almost no current draw on the input. Since no current flows, I suspect the tube is faster and better able to track the dynamics of the signal.
I appreciate these little videos. They're great!
I also know this is one of those grey areas which will never be resolved to everyone's satisfaction.
That little video is 10 minutes long
Great rundown on the differences. The generality is mostly correct. Enhancement Mode Field Effect Transistors do have very low input current in the DC area and can be operated with electrostatic charges. The difference here between them and vacuum tubes is a grid in a vacuum tube has very low capacitance as it is thin wire physically spaced away from everything else. In a Enhancement Mode FET, the gate capacitance is quite high, and even worse changes with voltage as the depletion region changes distance creating a distortion that tubes don't have. For this reason, FETs have to run with relatively high amounts of negative feedback and do not sound the same as tubes in low or no feedback applications. MOSFETS are best used as either switching devices in inverter power supplies such as automobile high power amps, or as output transistors with high levels of feedback to control linearity.
Bipolar Junction Transistors, have various amounts of linearity depending on the device. Some are very linear and make great output stages in a classic AB biased push pull output stage with little feedback for linearity. The family of curves on these transistors can be seen when driven on a vintage Tektronix curve tracer. Some have excellent linearity over a family of current steps.
www.valuetronics.com/product/370a-tektronix-curve-tracer-used?gclid=EAIaIQobChMIgsi3497s3AIVj19-Ch2jpQLQEAQYAiABEgKBFPD_BwE
Our shop has one. Unfortunately the voltage step range is often exceeded by more modern MOSFETS with higher gate voltages. It would be nice if they updated the instrument to include modern MOSFETS. Some MOSFETS require more than 3 volts gate to begin conducting.
3 different brands of the same tube sound different .
Nice explanation Paul. Anyway, yes they sound differently. But the main reason are the output transformers with their hysteresises and the slew rate. Due to the high value of the tube setup, they sound differently and gives you the warm feeling only tubes can give you. Signalwise on the paper go with BJT or with FETs if you like.
Howdy. I'm with you.
In my mind. Also the magnetization of the transformers play a role. It is not linear.
Regards.
The disclaimer at 7:00 is pure gold. Great video. Thanks very much!
I thought I knew a lot about audio and now I learn something new!
I researched Tubes and found that a scientific comparison had been done in a Lab. The results were most subjects prefered the music that was amplified by tubes. They said it was because of the fact that tubes have inherently a more natural distortion and that is more pleasing to our ears. So the report stated simply: The human ear preferes the distortion and final tones that tubes produce.
May we have a link to this study?
For sound to work perfectly, there's only one way to put digital to analogue ... But there have to be good speakers with a good twetter to work!
I like tubes. They keep me warm in the winter!
Interesting you mention that idea. I remember having a colour TV, bought way back in 1971, that had some tubes in it. It was in a small room and when it was on and the door closed it did a nice job of warming the room in the colder months of the year. I remember we would say our show is coming on soon, time to start up the heater!
Yep FET's and JFET's are very close to Tube smoothness ! That's why use them in my guitar overdrives. Transistors are Faster, that makes them great in front of a tube guitar amp.
The smaller Soviet tubes have very good lifespans (5000+ power on hours), so the "short lifespan" argument is not that distinct now (as well as these tubes being very readily available). There are a pretty large number of 6H3N and similar tubes appearing in the very modern range of hybrid amplifiers, and because the power dissipation of the "pre-amp" stage tubes is low, their lifespan is obviously far greater than that of power tubes (e.g. B300, KT88, etc).
yes tubes wear out but no one says you have to buy a tube amp with a large number of expensive tubes. i use a musical paradise set tube amp that takes all types and finding 2 new output tubes every couple years is not pricey and i find it fun. plus i can alter the sound 'quality' by using different types.
I bet much of it just comes down to size, as it takes longer for electrons to travel the length of the tube compared to the tiny distance of the silicon, you get different behaviour, for example, to electrons already in transit, when you charge the grid... so it's like the difference between a pneumatic system and a hydraulic system, where the pneumatic system softens changes as the gas can compress, whereas the incompressible hydraulic system responds very suddenly
Tubes sound different but they also measure different. You can of course claim something to sound different without measuring different just by not measuring what makes the difference. Tube amps are usually clipping the peaks softly and you get more musical distortion (even harmonics) than some hard clipping transistor amp. A simple FFT measurement or square wave comparison can easily prove this difference.
In the guitar amp world. There are a lot of units that use a tube preamp, and are solid state for the power section. A good 12ax7 triode-balanced tube will last a long time. A good tube like a Tung-sol etc, really adds to the preamp section in that saturation is achieved if pushed, as opposed digital clipping . That’s my 2 cents for the day ;)
Each sounds good in its own way. Manufacturers of SS and Tube gear have successfully bridged the gap in differences in the quality of sound. The only real difference left between the two types of designs is that tube gear lacks the juice SS does when cranked up which is a serious issue for me.
Hi, I only understood that they are different. But, what is difference in sound between tubes and transistors, are tubes better at providing low frequency response?
Tubes are vacuum so they won't generate noise and reproduce true input signal due to very high input impedance (if you don't overdrive it), the only drawback is output transformer that dictates the frequency responds.
I hope I don't sound unpleasant but it doesn't sound as if this Paul fellow really understands tube electronics deeply enough.
sir are the vaccume tube sounds loud and clear like the qsc rmx series??
Sir
Thank you for your knowledge and patience to try to explain the the very complex theory of electronics and sound
You are in my opinion the best of what UA-cam can offer
Thank you for what you do
I’m truly a real fan
Good video explaining how tube amps work in the general sense. Though, I was distracted by the chirp of the smoke detector every 30 seconds or so. I kept counting the seconds till the next chirp.
There will always be tube amps that are made to sound “tubey” and transistor amps that sound “transistory”. From what I can hear though, the best of tubes and transistors are sounding more alike. As they get closer to the ideal they are converging toward each other as well. I do think the ultimate synergy will probably be the tube input-transistor output approach. Each device doing what it does best.
Nothing can beat the vacuum of space, would you still need the glass casing around the filament then, Or is that not how it works?
The real difference between tubes and MOSFETs are the input capacitance. And the internal impedance due to the load. And not to forget - tube amplifiers needs an output transformer to match (e.g.) an 8 ohm speaker, with the limitations (or advantages) hence to this fact.
I'm with you! Years of thinking, no preamp is better than a preamp, i bought a pass b1 (and it don't amplifies)
and power amps got second ranging.
Using a (in Germany Dynavox ET 100) Chip Class A/B Amp, it's as smooth and musically as the Musical Fidelity A1 Integrated, without the Class A power consumption.
Also the dynamic is improved remarkable.
Thanks for your Clip, i enjoyed it .
What about the difference in distortion? That is the primary reason WHY they sound different. Tubes produce even harmonics whereas transistors produce odd harmonics. Also, transistors have hard clipping whereas tubes have soft clipping. Their frequency response curves are also different. That is what gives tubes their "warm" sound.
Operating voltage range doesn't sound like a major reason for the sound differences. Tube amplifiers sound different because they have primarily second harmonic distortion (less harsh than transistor distortion), high output impedances (less damping on speaker resonances), commonly use single ended outputs, typically operate in class A mode, use transformers in the output stage and have modest feedback (if any).
Reports from people who have better ears than I do suggest that tube amplifiers are hard put to compete with a good class A transistor amplifier - or even a basic one like the J Linsley Hood amplifier.
Obviously they sound "different" but a better question is: When the goal is purely linear amplification of an input signal (not intentional distortion like a guitar amp), which does so with the most faithful recreation of the input signal's characteristics?
germanium tranistors had a tube like sound. my magnavox receiver has germanium transistors so it has a warm tube-like sound
Think less is more. What worked in the past works today. I am a converted tech head and realized the error of my ways. There is nothing like a good LP through a tube and some great speakers. It's a ritual , it pays homage to many decades of refining the design and it sounds natural and musical. There were engineering Giants in sound in those days. Newer seldom means better in the world of hi-fi.
It still needs to increase gain of and more high-pass filter of your microphone.
My thoughts are that the human brain has the ability to determine if the input its getting is real or a facsimile.
I think this is the core of the tube or transistor, analog or digital debate.Analog audio and tubes are a closer reproduction of the original ,even with its technological flaws.
Tubes and Bipolar Transistors sound different because they operate by completely different principles, the word "Transistor" is short for the devices full name "Transfer-Resistor".
They are not so different as you might think, not when you get down to the electron level; its still an approximation of a little guy opening a gate.
@@suzesiviter6083 Yeah true, but tubes are mainly voltage-amplifier devices whereas solid-state transistors are mostly current-amplifier devices, yes there are Fets (Field Effect Transistors) and MOSFETS (Metal Oxide Semiconductor Field Effect Transistors) that operate very similar to the way a tube operates.
I’m sorry everybody but, when my 45 watt Solid state guitar amp was in the shop 40 years ago, I borrowed a buddies 6 watt Fender champ. That 6 watts of tube tone blew my 45 watt SS watt amp out of the water! I ditched that thing, used tube amps ever since and never looked back. So your Line 6 “models” a Tweed Bassman? Evidently, you’ve never played a Tweed Bassman!!!
Jack Dixon: I dont doubt it, but you know instruments like the guitar can sound better with added distortion, so this doesnt really apply to Hifi where you want the sound to be transparent.
@@suzesiviter6083 All good tube amps are super transparent, no audible distortion, what they actually call ''distortion'' is the non fatigue, sweet sound of tubes! Actually music sounds more real and live on tubes, especially jazz! But each to their own...
@@HASHEAVEN A good DSO will prove you wrong, Valves simply cannot go into the low THD figures a good bipolar or MOSFET amplifier produce; you prefer the sound and that's fine; but it is false to say you cannot hear the distortion relative to a good transistor design. Valves produce different types of harmonic distortion; this is one explanation why some people prefer Valves; IMO the warming up time and decreased life of tubes negates any small sonic advantage-but each to their own.
@@suzesiviter6083 I also like certain solid states like the simaudio moon, McIntosh, Accuphase and the clarity and detail of passlabs but those are all out of my reach. Maybe your are right but I certainly can't hear any distortion with my line magnetic, which I think sounds clearer and more detailed than many SS at it's price. Maybe what people call tube distortion is just the characteristic tube sound that I love...
For ex I don't like the sound of an old SS Rotel receiver that I have, or a technics mosfet and a marantz of a friend and my LM 219 has better soundstage, detail bass midrange and more information on the highs than those, which for many people are good Solid state Amps.
@@HASHEAVEN Well whatever works for you, Valves do look sexy. I still have the old Pioneer A400 which was raved about in the 1980's; I haven't wished for an upgrade yet. Some of those old Amps can still keep up or better todays amps, the A400 for instance used custom silicon on the output transistors-which is rare on all but the most expensive amps even today. My biggest mistake was getting rid of my old Linn LP12; that had a certain magic and yes you can argue Vinyl has higher distortion; but the sound was just so much more enjoyable than a CD; so I know what you mean.
Valves are voltage-controlled devices. Normal transistors are current-controlled devices. FET transistors (Field Effect Transistor) are controlled in the same way as valves - by voltage.
I love to ear and see ... the PASSION from your videos ! thanks and keep it up !!
thanks sir for all of these videos, there is alot of myths around this stuffs and what is better than that to see someone with practical experience explain it, every time you upload a video i can see i learned alot from it.
In transistors & all solid-state devices, the conduction of electricity is by "holes" & "non-holes" (electrons) which produce a hell lot of noise. But in vacuum tube, it is thro' pure electrons.
simple, tubes produces even/odd harmonics adding to the signal bec. it operates to be heated to function (filament/heater connections - 6.3v, etc.), whereas transistor/solid state
device just needs some biasing in a fraction of a current in milliamps or millivolts. tubes plates voltages requires very high voltages (100v, 200v - or >1kv) defending on kinds of output tubes used in an amp and it's topology or design, i.e. single-ended or push-pull, class A, etc.....................
Being a tube guitar amp player of many, many years, it seems to me that tubes in the power section of an amp is what gives it its "balls".
I understand, to some point, how tubes have a "softer,rounder" clip in the pre amp stage, but these hybrid amps that have a tubes(s) in the pre amp, and ss power, sound just as bad as all ss.
Put a ss preamp in front of a full tube power amp, and you are mostly there.
Of course, all tube is best, but man i am getting too old to carry around a 95 pound 2x12 combo.
I would love to find a ss amp that actually delivers.
Not really a valid argument. most guitar amplifiers are designed to colour the sound. They are a part of the overall instrument. Seems to be some confusion as to warm reproduction and accurate reproduction Not possible to have both as the warmth is distortion. it might sound good but it aint HIFI
I was at Fin's old gas station spot in Baker when those wings came in. Helped drag them across the yard
Perhaps I missed something. I'm 66 years old, have owned nice hi-fi equipment most of my life. I didn't get how they sound different from one another? What exactly is it that makes them sound different? I know tubes are supposed to have a "warmer sound," maybe "smoother" or "rounder" are better terms. What are we hearing precisely?
I always said the control signal in tubes is higher than the audio signal. In solid state the control is lower than the audio signal. So when you pump up the volume on ss devices, the audio signal seeps into the control, then you get bad clipping.
Is there a significant difference between using a 12 meter 16 gauge silverplated audio cables vs 11 gauge ofc copper cable that is short? despite the resistance which difference is 0.3 ohms which probably isnt even noticable and wont cause any signal loss maybe a small percentage of power loss right?
Tubes in the winter and chips in the summer - that’s my deal. I like the sound of both.
I think you said even if the data shows the plots the same they still sound different and if so I find it hard to believe. I am curious if you ever did a blindfold test lol and how you did? Would be interesting. I use mosfets to pwm control motors so was wondering if you do that with speakers? If so you could linearize if needed or closed loop servo as well which is what we do. Or you could make a lookup table to replicate the tube response. Would love to know. Thanks for the video!
Valves/tubes often use push-pull transformers which is where a lot of the "bad" distortion is cancelled.
This is almost always overlooked.
Howdy. One possible explanation.
In a BJT the collector current is dependent only of the base current and the gain factor. Collector voltage has no effect on the collector current. Within usable limits, of course.
Tubes are different. The anode voltage will have some effect on the anode current. The gain factor is not a constant. The gain factor will be smaller at high signal when the anode resistor causes the anode voltage to be lower. This is an inherent compressing property of tubes. It introduces the third harmonic to the signal.
Regards.
Research "Early voltage."
@@marianneoelund2940 Hi. Blimey.
I stand corrected. The collector current is truly slightly dependent on the collector voltage. Thanks for the correction.
In high regards.
8:25 "anybody who will tell you that [three devices] ... running at the same [specified parameters] ... do not sound different because they don't _measure_ different ... is [wrong]"
You never did explain exactly how or why in detail. I'm still not sold on the argument that tubes > transistors.
But my focus is on the "measurement" itself. If instruments have finer precision and better resolution than the human ear ... and they "measure the same" ... then how can the things they measure "sound" different to the human ear?
Exactly.
The thing is that most music producers will make make it so the sound that we ear has tube sound being played by solid state amps... which renders all this discussions immaterial.
Tubes produce 3rd harmonic distortion which gives that 'warm' sound
what if they made the filament from a metal combined with tungsten ? it might make the tube last longer
Paul's explanation is somewhat incorrect. Tubes only function in a vacuum because air blocks the electron flow between cathode and anode/plate (beyond the filament burning up). To my understanding, filaments usually are tungsten or an alloy of tungsten and other elements, and the cathode (the part that actually does the thermionic emission/"boiling off of electrons") is made of a combination of alkaline earth metal oxides (e.g. barium oxide or strontium oxide).
rationally cool description of hot tubes in right places (although I vote for precision opamps too), tnx ))
nice video ! no wonder the dogs a keen listener and paying absolute attention ! :)
Tubes do sound differently. I have a 1957 The Fisher model 80T Hi-Fi mono AM/FM tuner and preamp. In addition to treble and bass control it has four loudness levels. I connected it to a solid state integrated amplifier. Whenever I want to listen to old hi-fi jazz or blues recordings, I use the 80T. The sound is far richer and warmer than that of the solid state own preamp, although it is mono. The technicians of The Fisher went to great lengths to have a hi-fi preamp with very low noise floor and great Dynamic Range. The four levels of loudness allied with the treble and bass controls provide the best audio configuration. This 80T preamp and tuner also came with a power amplifier which also has it's own four level loudness control. I haven't used yet, because I have to figure out a way of commuting the speaker boxes from my solid state strereo equipment to this mono power amplifier and vice-versa.
Great Videos Paul! You have cleared up some understanding of gear and how things work! Thank You for making these videos. If i get out to CO. I will be paying yall a visit!
Start again from basics, please.
A bjt is a linear current amplifier (current in-current out, small input impedance, large output impedance), a triode is a linear voltage amplifier (voltage in-voltage out, large input impedance, low output impedance), a jfet/mosfet/pentode is a linear transconductance amplifier (voltage in-current out, large input impedance, large output impedance).
A loudspeaker driver is a motor, therefore driven by current (to behave linearly). But then, there is the loudspeaker crossover circuit, which is usually designed as a voltage devider, unfortunately.
From these basics, add the high voltage bias of tubes (that increases electric carriers mobility) and tubes' "excess" amount of carriers (as electronic cloud inside them) that helps in soft clipping at large transient demands.
From the above all combined, go to amplifier's output driver stage, then to the input stage and finally combine them all together, keeping in mind that we are used to design (from the beginning) voltage mode circuits that are easier (in controlling voltage). We are also used to voltage signals, but that's another long story.
Have a nice day, all.
Not quite following you Yiorgis Alexakis. By triode do you mean vacuum tube or valve triode? Because tubes have high output impedance, which is why they typically use output transformers to connect to loudspeakers.
Stuve715, you refer to "high impedance" in relation (relative) to the loudspeaker impedance, which is quite lower. Driving a relatively lower impedance, is what makes necessary the use of output transformer (and that does not contradict to the "low output impedance" concept of a voltage amplifier).
In general, a triode tube/valve- being a voltage amplifier - should present a "low" output impedance (if it is a good voltage amplifier) to prevent action of a voltage divider between its output impedance and the load impedance (something that alters output voltage on the load, thus "violating" the constant voltage source concept of a voltage amplifier's output.
I started my comment by saying that going back to basics could explain a lot of vague impressions that survive in audio in general. For example, all tubes do not have the same amplifying behavior. Pentodes are not voltage amplifiers, as triodes. They are transconductance amplifiers.
Another one: higher voltage JFETs/MOSFETs are quite different from low voltage ones as far as transconductance linearity and dynamic range (but that is another long story).
I wish you did the video, and explain current amplifiers as I always think volts is the energy difference that causes the current, can't think of a flowing current causing a voltage. Also, I thought you want high impedance i/p so the volts at i/p not affected too much ? Also isn't it something to do with harmonics why valve clipping is better ?
Starting from your last question, harmonics is a phenomenological result (that we measure) of what is actually happening inside the tube and that is the excess of electrons stored as a cloud and the current demands they can fulfill. These electrons contribute to soft clipping and also when a tube is over driven. We refer to the larger percentage of even-to-odd harmonics of the tube's output to just distinguish the perception that even harmonics (governing tube's output) give "light" to a more acceptable sound to our ears (odd harmonics are sounding harsh to our ears). Harmonics are produced from the non-linearity of an amplifier's output current-to-voltage characteristic (abrupt non-linearity produces more odd harmonics), which also has to do with the availability of carriers demanded by the signal to be driven to the amplifier's output load.
Now, as far as high input impedance, it is indeed necessary for avoiding voltage signal losses at the amplifier's input. It is a "must" when our input signals are voltage signals, especially small ones. Higher input impedances can be found in amplifying devices driven by voltage, such as voltage amplifiers (e.g. triodes, opamps) and transconductance amplifiers (e.g. jfets, mosfets, pentodes).
As for the current amplifiers, they need to be driven by current at their input as their input signal, in order to behave as linear amplifiers at their output. The device example we have is the BJT.
Finally, as you mention, voltage is the potential difference that is needed for the current to flow in between. But "free" moving carriers are needed in the first place for the current to be created. This is general and it doesn't contradict with the case of current amplifiers, at all.
I hope I did give you at least a rough idea about the subject. The limited space here is not suitable for more details. You can find many resources about all types of amplifiers at the internet to continue your quest.
Thank you very much for answering my questions, I can see that you have a great deal of insight in this subject, and I will look into BJT current amplifiers more now, thank you.
Thank you for all of your most informative videos. Wonderful stuff you share. Very grateful!
Thank you for sharing, very interesting and informative, just uploaded a few of my hand crafted tube Amps. I enjoy learning more about the world of tube Audio!
A tube amp and transistor amp should sound identical if well designed, equal output, and operated within their linear specifications. If you start to drive them out of non-linearity, all bets are off.
So Paul, how many hours of play do tubes last before they decline in sonic output.
Tubes DO typically sound different... because they introduce additional nonlinearities into the system beyond what a FET or BJT amplifier would. Tubes warm the sound by distorting even order harmonics and allow for "soft clipping" by a similar mechanism.
You can just about digitally emulate the "tube sound" at this point, but it's still probably worth it for studios to record with actual tubes. However, for playback I strongly prefer a transparent sound... which typically means solid state.
tubes also produce un-even harmonics. at certain impedances they produce even more uneven than even order harmonics.
The tubes cushion the sound, that is, they dampen it. I don't think it's any simpler than that.
Why do they sound better? I'm not an engineer or an electronics technician, but starting from the fact that when an electrical signal reaches the tube it has to be converted into electrons and pass through a gaseous medium, perhaps it is this medium that adds dynamics to the flow of electrons.
One interesting thing I have noticed empirical.
If you replace one BJT with another in an amplifier, the sound will be more or less the same even if the transistor is of another brand and have other specs( within reasonable limits ).
But if you replace a tube with another one, the sonic imprint usually is totally changed.
That is what I call interesting. And a bit of a mystery.
vacume tubes surley will be sounding different every second they are played . Or am i wrong and they wear out in stages. Come to think of it it dosnt even need to be used as some deterioration is happening due to contaminents that remain in a non perfect vacume??