| Revision as of 21:52, 7 January 2007 editTubenutdave (talk | contribs)496 edits →Tubenuts replies← Previous edit | Revision as of 21:53, 7 January 2007 edit undoLight current (talk | contribs)30,368 edits archivingNext edit → | ||
| Line 1: | Line 1: | ||
| ] | ] | ||
| ] | |||
| == Dummy heading == | |||
| '''Readers should note that this entire page seems to have been editted by Light current under his unilateral opinion that salient replies interleaved in context a somehow a problem. | |||
| *He has in places left his onjections in line, but moved many of my replies to a collection lower down, seemingly leaving his objections unanswered. | |||
| *He has rearranged things such taht the indetation is all messed up and it is thus impossible to follow the original sequence or attribution | |||
| *he has rearranged things so that text now appears ina different context from which it was originally written | |||
| even I as the original author of one side of the discussion below can no longer make sense of what relates to what | |||
| Needless to say my view is that his actions here are lacking objectivty. make your own mind up] 21:42, 7 January 2007 (UTC) | |||
| ''' | |||
| ---- | |||
| Information repeated from a cited reference published in 1976 (and presumably originating even earlier) as the main page does, is imho of questionable "serious" interest, except to historians] 16:59, 1 January 2007 (UTC) | |||
| maybe you should before going any further with this discussion ? In a high end System ? | |||
| :Why when I ve said Im prepared to beleive they sound different?--] 03:37, 3 January 2007 (UTC) | |||
| To suggest that tube amplifiers costing in excess of $50,000 .. and hailed as amplifier of the year (citing the Audio Note Ongaku as an example) are somehow "defective" is imho somewhat unreasonable, and possibly arrogant coming from someone that hasnt heard a tube amp for years and maybe has never heard a really high end tube amp playing in a high end system.] | |||
| :Where did I say that?--] 03:37, 3 January 2007 (UTC) | |||
| Aside from being another sweeping generalisation, This statement doesnt even make sense. | |||
| Do you mean open loop or closed loop gain ? Most transistor amps have very high open loop gain and if fed with a line level signal and with the NFB loop removed would simply blow the speakers to bits with a near sqare wave signal orders of magnetude beyond clipping.] | |||
| :Exactly--] 03:37, 3 January 2007 (UTC) | |||
| :Exactly what ? "an X that is working as intended is better than a Y that has been broken" ? what does that tell us (that is meaningful) about anything ?!] | |||
| More to the point there are very very few transistor power amps designed to be used without NFB. The only ones I can think of at this point are designs from Nelson Pass. For the record (I am more open minded than I think you have assumed) Actually I built and still have a (water cooled !) Class A "son of Zen" amp designed by Nelson, Thats a No Feedback single stage fully differential Mosfet amp, running in class A. It sounds wonderful actually. And very like a tube amp. Unsurprisingly because the circuit topology is very reminicient of a differential pair tube stage, and MOSFETS have a number of similariies with triodes. As noted in the original text I wrote on these pages some years ago (which you deleted long ago) the "classic sound" of tube amps has as much to do with the topology of "tube circuits" as it does to do with tubes (and ditto for transistors.) There are good transistor amps and bad tubes amps out there as well I freely admit.] | |||
| This page is supposed to be about tube amps - good or bad. This page is imho not a place for you to "vent your spleen" (to quote your own comment) about if tube amps are inferior or otherwise to transistors. Or ? Tubes amps ARE. Transistor amps ARE. There are many, there have been tube amps at least for over a hundred years, and many people like them, and some are interested in them. Isnt that enough to justify a Wiki page ? I am not interested in what you prefer, consider better - nor do I seek to pursuade you. You can listen to what you like. Thats not a reason to deny or undermine a wiki page devoted to trying to describe and explain tube amps to anyone that whats to learn about tube amps] | |||
| :Where have I been venting? Ive said few words compared to you.--] 03:37, 3 January 2007 (UTC) | |||
| QED ?! If you accept that transistors are inherently unlinear, you then rely on something else (usually NFB at audio frequencies) to lineariese them. Now NFB can only reduce errors, never remove them. and it can only reduce an error after the error has occured. NFB can thus improve the linearity opf a closed loop circuit, but perfection is imposisble to reach even in theory based on this approach. In reality there are many other problems it introduces (refer Colloms)] | |||
| never said it was possible. What I did say is that tubes (especially triodes) are much more linear open loop than transistors. To the point that tube amps that sound amazing are routinely designed using dramatically lower levels of NFB that is normal for transistor amps. Indeed some have no NFB at all. I also noted that NFB is itself imperfect.] | |||
| If you take something that is grossly imperfect (eg a transistor), you will never get to perfection by trying to fix it with a mechanism that is also flawed. Since NFB can be applied to tubes as well as transistors, starting of with something that is "less broken" to begin with has to be a good thing. And before I get accused of bias again, yes transistors have many other advantages (notably they are cheaper. but a few others too)] | |||
| agreeing that perfection is impossible with tubes just as much as transistors, with or with NFB, is NOT a reason for abandonning the PURSUIT of perfection. Indeed the whole game for amplifier designers and engineers generally is to "make a BETTER XYZ". even if better is still less than perfection] | |||
| === NFB place in audio amps === | |||
| OK you dont believe NFB has any place in audio amps!--] 03:37, 3 January 2007 (UTC) | |||
| :No, you are grossly misquoting me again. I have stated and restated endlessly that NFB is widely used in audio and for good reasons. That doesnt mean it is perfect though, or that there are negative consequeces to using it. Audio design (as engineering in general) is about design compromises, and seeking to maximise the balance and spread of positives while minimising the negatives. ONCE MORE for the record, notably NFB is useful in tube amps as a way to minimise (especially) crossover distortion in class AB / B amplifiers, and (generally, even in class A amplifiers) as a way to reduce Z out and thus improve the damping factor. But many will say the sound quality suffers in some respects. The price you pay. Why is more contravercial/complex but the fact that the group delay of any real world feedback loop in non-zero seems central to some of the arguments] | |||
| And to say that ''all'' valve circuits can be designed without NFB is plianly nonsense. For instance a cathode follower has 100% nfb applied.] | |||
| :(a) I never, ever wrote that ''all'' amps should be used without NFB - In the same way that I have never, ever stated tube amps were <sweeping generalisation> better than all transistor amps - your ire has taken you to the point that you are simply inventing gross exaggerations in order to misrepresent, please stop] | |||
| :::?! "to say that all transistor amplifiers are green and made from mushroom soup if plainly nonsense" .. "just as a thought" .. is rediculous ! Why write such a thing ? | |||
| :::more seriously, As you wrote the originally comment it gives the impression that I was alledging such a thing. We should take care not to mislead. or even to write things that could be read in a way that misleads. ;-) | |||
| :(b) for the record however, many amplifier circuits are largely designed as stages, each sans NFB, just as part of a loose overal concept that has a designed overal open loop gain that is then "wrapped up" inside a chosen global feedback loop to give the desired closed loop gain. (more recently there may be one or more nested feedback loops each enclosing more than one stage)] | |||
| ::Yes but many gain stages have local feed back. eg cathode or emitter follower. Or the ultra wide band 'shunt-series' pair and of course the opamp, or transistor simulations of it.--] 03:37, 3 January 2007 (UTC) | |||
| :does not sound too promising as an example to what serious high end gear is capable of .. We should not dismiss radio out of hand based on a cheap battery powered radio I had as a kid. Nor should we dismiss hi end tube audiophoile equipment if you havent heard any] | |||
| ::I didnt say you did say ''all'' amps should be used without NFB Im just listing my thoughts.--] 03:37, 3 January 2007 (UTC) | |||
| ::I didnt say the amp I had was any good! I was just saying I had one!--] 03:37, 3 January 2007 (UTC) | |||
| :Indeed sometimes during testing (or fault finding) the feedback loop is removed. But this is all irrelevant in any case. Any amplifier should be used the way teh designer indended it, to add unintended / extra NFB around an amplifier may well make it marginally unstable or even turn it into an oscillator. To remove NFB from a holisticly designed complete circuit that deliberately factored in additional gain (inevitable at the expene of some other compromise) in order to provide / increase the number of dB NFB that can be supported at a given closed loop gain invalidates any right you have to any given performance level. If you want to fiddle with the amount of NFB, then you must take on yourself the responsibility of being a designer, and as such carry the critism of your designs if they dont perform / sound any good. Its harder than it looks, there is much more to designing top class amplifiers than being able to calculate a load line and bias a transistor | |||
| :(c)When discussing NFB most people are discussing externally applied feedback (eg global feedback around > 1 stage) rather than what is sometimes called "degeneration", (or indeed simply biasing) based on ohms law on the resistors (including the internal equivalent resistance within the tube) on the stage itself. Although I agree eg a cathode follower is a FORM of feedback. The important difference is that degeneration does not have any associated group delay / phase lag etc. Whereas global feedback does (and this is the start of many of the problems) | |||
| :what ?!. It was possible to get very linear tubes even in the 1930's. VERY linear. Checkout eg the 845. I dont see your point at all.] | |||
| :Who was ever saying otherwise ? But you miss the point. Music isnt about a "continuous carrier" sine wave of constant amplitude. Its ENTIRELY about transients. In the same way that makinga decent sports car isnt about going fast on an infinitely long dead straigh highway, its about handlig on twisty even three dimensional country lanes. Failure to grasp this (because for them its not an issue) being the reason that the Americans struggle so much making decent sports cars ;-)] | |||
| ::OK but tubes not perfectly linear- are they? So they give distortion!--] 03:37, 3 January 2007 (UTC) | |||
| and if well designed can give reasonable ]s also.--] 03:37, 3 January 2007 (UTC) | |||
| I think its probably possible to design a transistor amp that has the same nonlinearities as a valve amp.--- it should then sound similar.] | |||
| :You might equally ask, it is probably posisble to design a transistor am that approaches the perfection of a valve amp ? ;-). ] | |||
| ::Depends what you mean by perfection. You could probably make the transistor amp have similar distortions.--] 03:37, 3 January 2007 (UTC) | |||
| :::We have been around this loop many times already. Dont confuse the distortions due to a given circuit topology (using class AB1 intead of class A, using a differential circuit instead of SE) ... which will profoundly define the distortion spectrum .. with the characteristics of the devices. Actually (as stated before) there are many similarities between triodes and MOSFETS, and between pentodes and bipolar transistors. (which is why its so damaging having valve sound as a separate page, BTW, much of what many people think of as "valve sound" is actually "the sound of the ciruit topology", not the valve. Otherwise you wouldnt be able to (approximate) it using a transistor circuit would you ?! | |||
| :::But it is only an approximation. There is a range of mu's available over them nay tube types, but generally triodes tend to the lowest gain and most linear. The "point" many auidophile tube enthusiasts are making is that "if done properly" (ie without mistakes or cost motivated compromises), a triode based audiophile amplifier can approach perfection (for selected values of "perfection") sufficiently closely that it isnt possible for a transistor circuit - any transistor circuit - to replicate its performance. Chainsaws are great, and very cost effective. But there are limits. A chainsaw will never be a substitute for a scalpel in the hands of a trained surgeon] | |||
| :Transistors are efficient high gain and cheap. Where in that sentence do you deduce the idea they are "superior" ? You yourself admit tubes are more linear. That may be irrelevant when making eg digital circuits, but when you want to make a linear circuit - of which an audiophile amplifier is one of the most demanding problems - its generally considered a useful characteristic. I agree that advanced design allows modern transistor amplifiers to deleiver remarkable results given the compromised materials with which they are working. They are also infinitely easier to mass produce, easpecially cheaply. Please dont confuse the economics of the retail consumer market with ideas about absolute quality] | |||
| ::To answer your comment,As noted above, yes you can build amplifiers with similar topologies using triodes and FETS (or Pentodes and Bipolar trnasistors), and there will be certain similarities in the results. for example check out Nelson Zen family on the web. Or look at the circuit of a Quad 33 / 303. But the differences in teh devices tends to mean that each generally does better when used in teh topology that over 50 to 100 years has been found to best suit that type of device. Of course. Engineers are not dumb ! Modern transistor amsp tend to direct couple - caps are awful. Direct coupling tubes is problematic. You can put a horse in an icecreme van, nd icecreme in a horsebox. But its usually not recommended] | |||
| :ROFL ! Whatever makes you think cleanliness = godliness when it comes to MUSIC amplifiers ? Do you think the rolling stones would have been a better band if they had worn matching suits and had nice clean haircuts, been polite and watched the conductor? | |||
| :why do you keep repeating this idea that transistor amps are somehow automatically cleaner ? Go to a studio and ask someone about amplifiers for capacitor microphones. Tubes rule. They say, not me.A triode in class A is the "cleanest" (most accurate, purest .....) amplifiying device there is, especially for small signals. Bar none, regardless of cost. If you do not know this you need to do some research.] | |||
| === Tubes with no NFB === | |||
| Why do you say tubes with no NFB are cleaner?] | |||
| :What?! For a start please will you stop equating class A with no NFB ? they are completely ortogonal design techniques. You can use, or not use NFB in class A or class B.] | |||
| :what I actually wrote, see above .... Class A (strictly class A1 if we are going to get uber picky) *IS* cleanest, by definition, because the gain device (NB I did not say tube) is operated entirely and continuously within its linear region. whereas in class B they are not] | |||
| :Regards NFB, this may be useful to reduce the amplitude of the distortion spectrum in th event that you have a design with poor open loop linearity, and some other benefits. However the price you pay for that in the real world is that due to the non-zero group delay of the feedback loop, information about signal transients is smeared in time domain. And since music is all about tranisients, that matters. Most successful tube designers since Williamson have settled for only modest amounts of NFB, in many case none, despite being well aware that increasing the open loop gain and thus the feedback margin would further reduce the measured distortion levels. Since the 1980's most designers of high end transitor amplifiers have also moved away from huge gain / huge feedback margins towards designes with greatly enhanced open loop linearity, and greatly reduced amounts of NFB. Because the concensous is that it sounds better that way. the sound is subjectively "cleaner" and more natural. (Even if instrumentation reveals significant amounts of predominantly even order harmonic distortion products since these have not been removed by large amounts of NFB)] | |||
| :QED | |||
| :At the risk of a GROSS oversimplification (sorry I havent the time for more) any idiot can make an amp with vanishingly small levels of THD (lost in the noise floor is usually considered "good enough") by simply making a direct coupled circuit with immense open loop gain and then wrapping that up in immense amounts of NFB. In the 1980's many people did exactly this - and AWFUL they sounded too. Today it is generally appreciated that while SOME NFB may have its place (for example it also usefully lowers Z out) it is much more important to produce a circuit that has at least *reasonably* good OPEN loop linearity too, but aiming for a much more complete "quality" than simply low THD after NFB has been re-applied. tranient response and stability are also critical. There is at least a loose correlation that very high NFB amps tend to have good measured THD and flat frequency response, but lousy transient reponse and sometimes questionable stability (especially into iinductive loads) | |||
| :FYI,Tube amps (even with mid level OPTs) typically have frequency response that is more than "flat enough" over othe audio band - even without NFB. Tubes will usually give flat response to teh hundreds of kHz or indeed usualy megahertz. The coupling caps are usually picked to have 3dB points in the Hz range, so a fraction of a dB down only at 20Hz. | |||
| :GOOD audiophile OPTs are sectioned to a degree that makes the works of art (look at a lundahl sometime) - far better than simply bifilar winding on a toroid core. Todoids are VERY difficult to wind the huge number of turns needed to give the wanted inductance for an audio OPT, C cores are normally used on quality products, although some excellent OPTs are even based on EI laminations. (refer Tango, Tamura etc). Toriods make more sense at high frequencies (SMPS, RF ..). But Audio amps are not like that. The core size and turns count you need to get >10 W at 5 Hz without saturating it is much bigger. You need to be able to wind teh sections tightly wit each wire side by side of the next, with no raggy gaps all over the place like you get on a normal toriod.] | |||
| === Editing === | |||
| I have the impression that you are a very experienced and very good editor, I do not question that. You may know a reasonable amount about electronics in general, Nor do I dispute that. However I get the impression from a few errors in some of your edits to this page, and from your statements above, that you do not have any deep knowledge about high end audio amplifiers, from either an experience (listening) or technical perspective. If that is in fact correct, and especially since you have stated that you dont regard this as a serious subject, maybe you should leave this subject to people who do have experience of it and do consider it to be very serious ?. Just an idea] 22:39, 2 January 2007 (UTC) | |||
| :Its info that people might reasonably expect to be in an encyclopedia. I would! 8-)--] 03:24, 2 January 2007 (UTC) | |||
| yes ! .. so long as its clearly labelled in the history section. Techniques for radio ham amplifiers in the 1960's are indeed factual information - but my point is that these are nt representative of modern practice for tube applications per se, ie all other applications. We should not mislead people ! ] 01:52, 3 January 2007 (UTC) | |||
| Im sure you can provide a great deal of help regards formatting etc, clearly you are much more practiced at edditing wiki pages than I am. However over an extended period you have taken down a huge amount of text that was relevant and informative and replaced it with text that was often technically imprecise, recycling misconceptions, or making statements that may be true in some cases in a form that make them seem to be universal truths, when the exact opposite may also be possible. The vast majority of the hours I have put into trying to develop these pages has mostly gone on back an forth trench warfare with you, each side endlessly clawing back to a position barely different from were we were before, when imho it should much more profitably for all have been spent developing the page | |||
| ] | |||
| === Circuits === | |||
| Many thanks.I agree that there is a long way to go this page and can be improved greatly with time. Actually I would *prefer* that circuits from one or more of the recognised cassic amplifiers (for SE, ongaku ?, for PP the williamson ? NOT the QUAD II imho, it was an iconic amplifier but has a quirky and unrepresentative front end). The problem is that I do not have the copyright to publish any of those designs here, if someone has a design they can publish instead, that would be fine ! | |||
| One "wholesale" reference would be the journal "glass audio", which is devoted entirely to audiophile tube amp construction, over the years it must run to thousands of articles by now. There is also a huge amount of on line material. Not least rec.audio.tubes .. however as noted this is a very religious subject and many adepts are self confessed fanatics, it is eassy to find people who believe that anyone using a triode with indirect heating should burn, directly heated triodes are the only true god. Dont even ask about using a pentode. a TRANSISTOR ?! GASP ! <drops dead of a heart attack on the spot> A schism within a schism within a schism. its hard to find common ground here. Here be dragons in fact (golden dragon is a brand of modern production tubes from china) ;-)] 01:48, 3 January 2007 (UTC) | |||
| Many (who HAVE heard good high end systems of BOTH kinds, including many HiFi journalists and other very widely informed and experienced commentators) hold the opinion that some good tube amps are much better able to capture the atmosphere, life and naturalness of music (esp accoustically recorded music) better than silicon amps - regardless of what the measured performance comparision is. Some DH-SETs soud amazingly good even at ? 1% THD. Some transistor amps that measure at <0.002% sound crap. Unless and until you can explain this, and present some other measurement that actuyally do give a correlated indication of sonic "acceptability" - many high end audiophiles would suggest that you pay only passing attention to measurements. Your ears are much more sensitive. And to a music lover, more important | |||
| FYI, I have also heard (indeed owned) extremely highly regarded (not to mention expensive) transistor amps, including from KRELL (KMA-100 class A monioblocks for the record). Most people would accept KRELL as being a reasonably good representative of breed - and equally I have heard them blown into last week by tube amps. | |||
| Well, ideally of course we want no distortion, we want perfection. Which kinds of rules out class B immediately. Sadly that takes will it a large chunk of mainstream consumer electronics. If someone WANTs some odd order or aharmonic distortion, thats a matter for them, but in that case I think we have nothing more to discuss with them about high end audio - they need help, and possible a hearing aid | |||
| :Whoever claimed they were superior ? I have throughout gone out of my way to cite the disadvantages as well as the advantages of both, and also to note that the majority of tube amps use NFB as well as (almost all) transistor ones.#] | |||
| :Didnt say you hadnt.--] 03:37, 3 January 2007 (UTC) | |||
| Actually you have. repeatedly. ;-)] | |||
| However such a statement as you make above is simply prejudice (not least as you admit you havent heard a decent tube amp in ages). Newer does not automatically equate to better. There are very many people who would agree with that statement (many of them also having never taken the time to actually try the alternatives they so enthusiastically condem). There are however also many people who HAVE heard who will say that transistors amps are not inherently superior. Wise people will probably find good examples and merits in both. ] | |||
| :I wasnt aware of making any prejudiced statement. Which one?--] 03:37, 3 January 2007 (UTC) | |||
| so many to choose from ...] | |||
| But the fact remains that NFB is not perfect either (I strongly recommend you read Colloms article referenced on this page), and at the high end there is a significant tendency to have greater open loop linearity and less NFB, with a significant slice of products using no NFB at all. The later being an option that transistors dont really give except in a very very few niche examples (refer my comments about Nelson Pass, below)] | |||
| :Yes it rules out class B if you prefer class A type distortion to crossover dist.--] 03:37, 3 January 2007 (UTC) | |||
| This is facile/misleading. The whole point is that class A gives dramatically less (asymptotic to zero in fact), and sonically "not so bad" distortion under small signal conditions, whereas class B gives exponentially increasing distortion of a very unpleasant kind under the same conditions. Anyone who would "choose" large amounts of unpleasant distortion over lesser amounts of less objectionable distortion imho should abandon any pretensions of listening to HiFi, which is after all about "high fidelity". NB this is not saying class B has no place or is "inferior". It is much more efficient and as such is the only viable approach for high power on a budget etc etc etc. But lets not delude anyone that from a distortion perspective it is on a par and "its a matter of choice which kind of distortion you prefer". Puh lease] | |||
| === POV === | |||
| I respect your pov! But its '''just your''' pov --] 03:37, 3 January 2007 (UTC) | |||
| :never claimed it was anything other than a POV, although there are very MANY tube enthusiasts who share it (and can give good reasons why, bot bsed on theory and what their ears tell them when they listen, and most of them are in addition to yube enthisiasts (thet they have BECOME) more generally Hifi and music enthusiasts, and they usually have very wide experience of listening to a huge range of equipment, of all types and all price levels. Of course there are also people with a pro silicon POV. I think it is interesting that the majority of strongly anti tube POVs seem to come from people with little or no actual experience of them. ] | |||
| :Whoever said otherwise ?. There is also a problem with what is "accurate" .. very low measured THD and a ruler flat frequency response does not automatically make something sound "natural". Yet of course the reference (reality) by definition does sound natural. This is where the fog starts, "good sound" is very hard to specify and measurements are very hard to correlate with the resulting experience.] | |||
| === NFB === | |||
| '''AGAIN WARNING to readers that the question below is not the point the following response was written to address, Light current has rearranged this is a way that my following text is now out of context and appears to say something that was never indended. | |||
| ---- | |||
| For the record modern op-amps are excellent ahve some have exceptional performance when used as an op mp is intended to be used. Discussion of whicg as no place whateseover in an article about tube amplifiers] 21:47, 7 January 2007 (UTC)''' | |||
| So is that why opamps perform so badly?--] 03:37, 3 January 2007 (UTC) | |||
| :groan. If we hit the last one first, an "op-amp" circuit is the classic example of an EXTERNAL feedback loop AROUND something (the "op-amp" being treated as a black box, you dont need to know or care how it is implemented internally, in principle) This application gives all the classical advantages of NFB, ie ideally the performance of the complete circuit is defined entirely by the components in the feedback loop, and not at all by the op-amps internal compnents or circutry - the op-amps only role is (ideally) to provide infinite gain, slew rate, input impedance and zero Z out | |||
| :regards local feedback. We could equally say that almost any biased transistor (one with a resistor between the collector/emitter and supply) has some *local* feedback (ie degeneration), just as much as a cathode follower | |||
| :what if "many gain stages have local feedback" (almost all, or some form !) What has this got to do with the price of milk, ie if external / global NFB is good (or otherwise) and why ?] | |||
| ::Where is the delay generated, and how large is it? ie what fraction of a cycle is it at say 20 kHz?--] 03:37, 3 January 2007 (UTC) | |||
| :::I havent got time to argue this endlessly. Refer to Morgan Jones, 3rd edition, page 408. Or most other introductory textbooks on electronics. You may like to consider how you can make a wien bridge oscillator out of a couple of caps, some gain, and a feedback loop. The Nyquist stability criteria is that any circuit that gives greater than unity gainat the 180 degree phase shift point will oscillate. OK ? | |||
| :::separately, as noted elsewhere, design of a audio amplifier does not stoop at 20 kHz just because human hearing has usually packed up by that frequency. Many poor designs oscillate in the ultrasonic range. Sadly the consequencesare very audible inside the audible range (quite aside from frying your tweeters.). Look up slew rte limiting and IMD . mixing | |||
| :Neither am I trying to convert anyone, or be biased. As noted I own and use both kinds myself. However I think on this page it is appropriate to try to explain the salient considerations behind why this type of ampifier, despite its many disadvantages (not least that they are inherently extremely expensive to make, whereas transistors are inherently very cheap) has so many enthusiasts. That is (and is all) I have tried to do - and do objectively | |||
| :?! please cite the things I have said that you consider to not be common sense, and I will happily try to improve the page to explain them so they are common sense (and/or cite some supporting evidence). However it isnt viable to have to cite a reference for every sentence written, nor is it desirable to have the article only as a large and disjoint series of citations from other places, it needs to fit together in a coherent and consistly readable form. More to the point as you have highlighted there is much more basic stuff still needed to be be written. When we have got a page that at least addresses all the major points We (or hopefully also help from others) can start to support the text with references. This is a huge subject. To explain everthing fully would take a book, not an article. I have cited a couple of books that do cover the subject imho completely and excellently in the references. Morgan Jones and Langford smith between them are some 1500 pages on this one subject. Glass Audio is several thousand pages more | |||
| :oh, thank you for you indulgence ! For my part, I will try to continue improving these few pages related to tube amplifiers, if you will refrain from gross tearing down of text that addresses matters you hadnt considered to be relevant to amplifiers due to your limited awareness of this subject as you have accepted above. I do not intentionally say that things are "Not common sense" once pointed out, but it is certainly true that audiophile amplifier design is very much concerned with subtleties that are far from obvious to people outside the field. This is NOT "general electronics". | |||
| :I have previously assumed that you are taking this personally based on the highly personal comments claims and accusation you have yourself written, and I would respectfully note that these same policies you cite here (which for the record I am well aware of) are equally applicable to yourself, and you are equally guilty of breaching them repeatedly imho. As noted elsewhere I became so incensed at your gross vandalisation and imho dmaging edits that I simply dropped out of trying to work with this for over six months | |||
| :I am happy to try to work together with you in a positive way on these pages, but I respectfully request that you give a little respect and question things that seem incorrect or inappropriate before you fire up the chainsaw and hack them out again. I will in that case try to improve the explainations of why they are in fact correct and relevant, or we can co-operatively figure out what to do for the best, yes ? | |||
| regards , Dave ] 23:47, 5 January 2007 (UTC) | |||
| ::Why are you not satisfied listening to the music as it was recorded? Dont you think the recording engineers know their job?--] 01:41, 6 January 2007 (UTC) | |||
| :::what ?! Where does this random comment come from ?. The whole point of HiFi is to enjoy the music as recorded by reproducing in the best posible way. | |||
| :::The accusation that I have anything but great respect for recording engineers is utterly unmotivated, utterly incorrect and unworthy of further consideration. It seems Light current once more inting is own view of the world] 21:31, 7 January 2007 (UTC) | |||
| == Tubenuts replies == | |||
| because you have repeatedly written statements implying that they are not simply "different", but are inferior to transistor amps. It seems to me as I read many of your comments that you have a negative opinion about tube amps, or at least are convinced of the superiority of transistor ones (for reasons you have never elaborated). That would be something we could discuss if your view was based on appropraite experience, or even theory. But to hold a strongly negative opinion about something you admit you do not have extensive personal or theoretical knowledge of suggest a hint of prejudice ? As you have cited several times, Wiki has principles requiring objectivity. | |||
| I personally go further and think its reasonable to have some idea about the subject you are writing about. recycling things you have read elsewhere may look formal in the sense you can cite the reference, but if you have not fully understood all the collateral issues surounding the text in its original location there is a danger that what you write may, after all, be less accurate than the original comment. One of the many interesting schisms in the tube amp construction world is that between western engineering mindset and the eastern existential mindset. In Japan especially, the DH-SET (which was dismissed as antiqued junk in the west) never went out of fashion, and indeed was continuously pursued further - they recognised that it actually sounded very good even if they in fact could not explain why at that time. That it DID sound good (actually better) was enough of a reason by and of itself. Many years later, western engineering and measurements developed to the point that they could start to see (and measure) things that were going wrong in the "modern" amplifier designs they had previously considered newer and therefore so much better (withou t actually listening it sometimes seems). Today it is possible to make very subtle measurements (if you have enough money) and the engineers have learned a lot, and today we can explain and understand a lot more. But understanding takes time. Really I urge you to try listening to a really good tube system if you hav the opportunity to do so - without worrying about why or how it works for a while. just close your eyes, relax and LISTEN. listening is actually a skill itself. the pace of life in the west has made people do everyting so fast yet superficially. BTW. listen. and enjoy. Then maybe you will have a better mindest to explore how and why the music that you hear is so very musical. cf "clinical", as many even very high end trasistor systems sound | |||
| Hummmm. I qould argue that actually it is extremely difficult to complete a successful design of an audiophile grade amplifier using either technology. The design has to be holistic if it is to succeed. OPTs are of course imperfect but that is primarily a problem for the designer/manufacturer of the transformer, it has little impact on the amplifier designer, other than to select his OPT maker of choice. the main design parameters (eg turns ratio & bias current and power handling) are not really affected by the make or even type (EI, C core, Toroid) of OPT. Its just as valid (and perhaps more so) to argue that transistor amps are much harder to design since the circuitry is much more complex, the parts count is much higher, and since they operate at lower voltage / much higher current (and are rarely class A fully differential) the POWER SUPPLY has much more severe problems with supply modulation, pull down and hence inter stage/inter channel modulation etc etc etc .. imho its grossly naive and misleading to make a sweeping statement that tube amps are "More difficult" because of the OPT | |||
| ::Perfection is even impossible with tubes. NFB or no.--] 02:06, 6 January 2007 (UTC) | |||
| :::of course. Perfection is impossible in anything. why waste peoples time with such statements. Perfection IS possible to approach, and improvements are possible. Just because one thing isnt perfect isnt a reason to settle for some thing ten or a hundred times worse. Some belive its worth seeking the best you can get. Many of thes people seem to feel that tubes are a good way to approach this goal.] 21:27, 7 January 2007 (UTC) | |||
| Please show me the : | |||
| <blockquote> | |||
| ''huge amount of text that was relevant and informative and replaced it with text that was often technically imprecise, recycling misconceptions, or making statements that may be true in some cases in a form that make them seem to be universal truths,'' | |||
| </blockquote> | |||
| So we can discuss it at length. 8-)--] 01:05, 6 January 2007 (UTC) | |||
| Please quote : | |||
| <blockquote> | |||
| ''text that addresses matters you hadnt considered to be relevant to amplifiers due to your limited awareness of this subject'' | |||
| </blockquote> | |||
| So we can discuss. --] 01:07, 6 January 2007 (UTC) | |||
| :Where to start ? ... how much more is needed? .... one could start by referring to the last 10,000 words or whatever in the history both here andon the user talk pages. Specifically as just one of very many examples and extended explaiantion on why thermal considerations are not a problem for tube amps. | |||
| :'''I am uninterested in being baited into further endless arguments that will just waste my time, it seems to me that Light current seeks endless argument. and responding to this will simply be futile.''' Im sure there is a wiki guideline to that effect somewhere. | |||
| :improving the page itself is more important and has a higher call on my limited time.] 21:27, 7 January 2007 (UTC) | |||
Revision as of 21:53, 7 January 2007
Talk:Valve audio amplifiers/Archive1 Talk:Valve audio amplifiers/Archive2