Vintage Hi-Fi Info
All contents of this Website are Copyright. Original research, photos of our hifi & all unquoted text is ©2011-2021 by select45rpm. This is all published freely on the internet by us to further the scene, not to give any seller or forum 'expert' undeserved credibility. We Do Not Authorise any Copying, Republishing or Quoting, even as rewriting Our Research In Your Words, of using or linking to any of our Sections on ebay, any sales sites or anywhere else. No-one else has formed these opinions, so don't steal them as yours. Please do not link to our site on ebay sales trying to use our unique info to play buyers for a sale suggesting we are authorising their sale, as we are not.
| *See our NEW Hi-Fi Blog page
lots of New Sections since Jan 2017 that add a wide range of Hi-Fi & Tech related subjects plus opinion on Hi-Fi News 1970-1980 as we read through.
Includes New Articles on this page's subject. This page has been updated & read through with 2017 ideas & with an Index to navigate.
Valve-Tube Hifi For The Best Sound...?
Everyone into Hifi should make an effort to try a Valve amp. Buy or Borrow a high quality new or fully working & restored old one. Only then will you understand what Hifi is about. Whether you want to keep one is another thing...
INDEX TO THIS PAGE: Valves Section ↑ TOP
INDEX TO THIS PAGE: ICs Section (needs some tidying...) ↑ TOP
Intro: It Sounds Valve-Like...? ↑
Transistor amps that are all original, ie not recapped or upgraded, are regularly said to be "Valve Like" even modern things stuffed with ICs which is impossible & just shows the seller's ignorance, as who has actually heard a properly done valve amp to really know? Vintage all-original valve amps may have a rich full bodied mellow tone, but the treble can still be ragged due to wrong Bias, high input resistors, poor limited design in all ways or aged components. There are different designs in Valves, read on for more on that. We've heard plenty of Valve Radios, Radiograms & Hifi Amplifiers both Vintage & Modern to know what Valve amps are like. Addictive. Valves are for thinking folk who can deal with them & maintain them. Valve amps are never just plug & play like a Transistor amp. Going half-hearted into Valves is a bad choice, buying a 50 year old one on ebay even if it's said to work will often mean it doesn't "not work" and it will often be of very poor sound & need a full rebuild. Don't buy one thinking someone will restore it for £300, the best we know you'll get it is "working" acceptably again. No-one will change the aged capacitors unless one is obviously failed, it's a Hobbyists task to keep valve amps alive. Or find someone like us to rebuild it for you, we can do it very nicely, but it'll not be cheap. See the Hifi Upgrades link at the page top. Compare it to restoring Vintage Cars & Bikes as on TV cable-satellite, we're waiting for a restore a Hifi show though 'Rick's Restorations' has tackled Jukeboxes, Early Radios & Wind Up Gramophones, you only see the external stuff. Would anyone want to watch us order capacitors after redesigning the circuit & soldering wires in? It's not terribly exciting beyond the technician yelling as bits go on the floor or vanish a second after you put it down, though the results can be the same you see delighters owners of junk get it back restored & prettied up.
This page has sections typed at various times with pros & cons of valves, some may repeat things a little until we re-read & edit it one day. But valves are considered the Ultimate way to the Best Sound & will better any post 1970 Transistor Amp, but lately our upgrades on some pre 1971 have bettered valves. The best Valve Sound is from a Modern Valve amp, the Vintage ones suffer from age...
A Lazy Reader opinion on Valve-Tube amps. You need to try Valves if Hifi matters to you, just so you know the sound & can go bore everyone by saying your IC amp sounds 'valvelike'. Even an old aged all-original one of 3w that crackles or hums will reveal the magic of valves & you'll be hooked & forever go searching for it. It's how we got hooked years ago after all. But the quest for the Perfect Valve Amplifier is almost A Myth as it doesn't exist. It can be created from good ones by upgrading as many DIY-ers & Guitar Heads have done for decades. But the hard truth is if you choose a 1950s-1960s valve amp it'll have a welcoming sound, but whatever you do, it'll never be as "hifi" as a modern one. We've recently tried the Prima Luna modern amp, a company that has just one design amid several amps it appears, but to plug it & play will be more agreeable than to buy a vintage one & hope to improve it, but neither are 'perfect'. We've tried, as this page reveals, but results can only go so far. Buying vintage, accept it's not going to be perfect, but enjoy the sound that very few transistor amps can deliver, despite fools saying amps stuffed with ICs are "valve like"... 'they don't know nothing' says we. Valves are for Mad People who like to dig deeper & to think a valve amp won't need attention or new valves will catch you out. Valves aren't cheap, valves are not for the masses, valves keep the room too hot in Summer: ah, but the sound... but only if it's designed right: that PrimaLuna amp was a limited design that sounded rough pushing the input valve too hard instead of adding extra valves. It'll impress many, but we tired of it fast. There is certainly scope in the PrimaLuna to upgrade it though, but the owner only wanted us to do more predictable upgrades, as going deeper into upgrading valves opens up a huge amount of problems that take much to overcome.
A Quick History Of Valves. ↑
Valves in Audio started in the 1920s, plenty online about the History & Invention. By 1925 Electrical Recording of records took over, Radios were improved from Crystal sets to valves & electrical horn speakers. By the late 1930s the type of Radios & Radiograms introduced in 1929-32 were commonplace. By the Hifi resarch progress of FFRR in WW2 furthered audio quality & NFB by the first Leak Point One in 1946 furthered Hifi. Things slowly grew, as most buyers of audio equipment bought a table Radio, Record Player or a Radiogram. In 1948-56 many ex-WW2 engineers started DIY audio as kits & parts became available & the first Hifi items as in the build-in type of preamp & power amp, all in Mono, started. By 1958 Stereo had arrived & by 1962 Stereo records were selling well. By 1968 the Mono version of a "Pop" record was outselling the Stereo one & by 1970 Electonic Fake Stereo was the thing on Mono tracks. Oh dear. 1962 saw the first Germanium transistor amps & 1964-65 introduced Silicon Transistor Amps, Valve amps were sadly in decline by 1967 & by 1969 hardly any were available. By 1971 Valves were gone in Hifi with only really Luxman keeping them alive & counting for 20% of their sales we read. But the Valve scene grew still with valve amps & the Marshall guitar amps & their like will have played a major part in advances. Most improved on, or just lazily copied, the Old Classic circuits with the quality valves the early to mid 1950s, the current valve scene is mainly based on improvements by Guitar Amp guys, bless you. By 1979-1980 a few brave companies were making Audio Valve Amps again based on the new ideas & by the early-mid 1990s valve amps were being made more often if looked on as insane by the majority, but who cares about them. By the 2000s the budget Chinese valve amps & a whole new scene of revived old Classic models made valves again what wise buyers preferred, tired of the ever-thinning transistor sound that started in 1972 & was the norm by the early-mid 1980s as each year the costs were cut even more. Fatman i-pod valve amps took valves to a new level, pity the speakers used with them will not do them justice. Valves today are easy to find as they were in the mid 1960s, but with new ideas & new markets wanting them. Old Valves for Radios from the later 1930s onwards are still findable, though the stocks of original EL34 & KT88 type valves are long gone. Valves & Vinyl Records too will never die. Unless you put your Pink Floyd LP on top of a valve amp & then find a crinkled work of art instead.
Why You Can Not Just 'Use' Vintage Valve Amps ↑
Simply because they will fail after years of not being used & likely be unrepairable. This we found out in the early 1990s when two items weren't even 30 years old. A Heathkit Oscilloscope we found fascinating for a few weeks, with our limited early skills back then we took it apart & it looked fine so was surprised when one of the transformers starting smoking & never worked again. Similarly a Stereo Pye Black Box, the shiny coffee table version that we'd known from the early 1980s, this one needed a few coupling capacitors changed to worl but failed badly too with the transformers smoking & the tar pitch melting off. Both trashed, if perhaps they could still be fixed today if caught in time. There will have been failed capacitors, resistors & everything else needs a full check to not risk disaster. So to those who buy Rogers Cadet III or any 1950s-1960s valve amps & just use it are risking the same catastrophic failures. Even on a 1936 Murphy floor standing radio, it never worked until we looked at the mummified capacitor block & put some new ones in & it played music again. A 1947 Ferguson radiogram hummed loudly as the main capacitors were useless. But if serviced & repaired-upgraded properly, even the oldest Valve Radios can work again.
Valves have the general opinion that they are warm sounding ↑
ie rich but soft on treble. This is nonsense on a properly designed, upgraded & maintained Valve amp, if it was true how would Capitol record Frankie in such hifi as valves were used for mastering. Valves, like Transistors & even CD sound quality, is reliant on design. Get it right & it can be wonderful. An aged amp can sound like that though. Many of the 1950s-1960s Valve amps bought, which usually means a Rogers Cadet or a Leak are now old & aged items that are past their best. They can be brought back to live by mindlessly replacing exactly the values of component for like, and wondering why it doesn't sound very good, or they can be upgraded revealing more of the sound than the original designs limited by low value capacitors & the need to limit to hide weaknesses elsewhere. Then the soft warm sound will be no more, but hardly anything old will be left beyond resistors & casework and you will still be bound by the old design. Valves designed properly can give a speed, a depth of bass fundamentals & a pureness of treble that transistor amps usually can't fully reach, though as we've found some early transistor amps can do an excellent job, the further we push our upgrades, we are surpassing valve sound which sort of changes things, to the point the valves get another look at.
The Ideal Valve-Tube Sound is... ↑
What all are searching for in both valve & transistor amps. The sound is a huge powerful rich sound but crisp & powerful on treble too, nothing dull muddy or warm. It takes high spec & the 'right' design to get this sound & be sure most valve amps can't do this sound & even less transistor amps can come anyway near. There are exceptions & only with intelligent upgrading. Deep powerful bass does not come cheap, which is why most modern valve amps still put in the 1950s spec low value coupling caps & still use the same spec as was in the 1960s ones. Progress is very hard with cost cutting & lack of open thought, many still copy the 1950s designs with no thinking to what could be bettered. The dynamics of valves done right will always be bigger than transistor amps, Marshall guitar amps are still valves for a reason. There are plenty of modern valve amps with low spec, lofty prices beyond their worth that still sound miles away from the Ideal Sound, but most don't realise the 'better' so are happy with them despite no Tone, no Phono or even a Mono switch. As for transistor amps, we've only ever heard one that truly sounds like a valve amp, even some of our 1967 favourites are only quite like valves, with the lack of the big powerful rich bassy sound. But if your valve amp is a bit thin sounding with a grainy harsh edge, you'd not know the true deep sound the best are capable of. But that, to be fair, is a domestic sound amp, the huge rich sound is great fun but more a PA type sound.
Take a Trip into Valve-Tube Sound Land... ↑
Written later than the section above, at a time possibly imaginary yet not forgotten & likely re-creatable if the Planets align similarly one day, here is a very different look at Why Valves Can Be The Best Sound. No shop bought valve amp is anywhere near this, now that would be silly, it's only for those who ignore convention & do it properly, taking months to get it perfected & taking more than a few HT shocks along the way, not all accidental perhaps. This was written using the Trio WX400U that we rebuilt, a lot. On headphones only will the best description be realised: again the absolutely huge dynamics is the key, even played loud you'll blue the air at how that note just hit you, just don't try to chase it. The unworldly Bassline is rattling your brain around your skull, steady on. The sound in mono is just so central in your head it's a little disturbing that you got 'that dead dude' inside your head. In Stereo, depending on how it's mixed, a lack of anything mid head but at a sort of / \ virtual shaped sound. A lack of boundary in the music pulling away all layers of wool from the sound, the sound you'd expect to hear sat in the studio as they laid down that track, not that you could be in all places at once as the mastered recording can be. The sense of space & depth to even Mono tracks will be easily noticed. No real need to say it's correctly focussed as that implies it was blurry & now tuned in, no it was never weakened in the first place. Effortless but unlimited in resolve it the deal here, no eek-a-mouse moments at rough sound, only deep wallowing in audio hitherto unheard if much imagined. Not dissimilar to plugging the Phono jack deep into your Noggin Box, but you try find the right socket... The best Hifi always lets you feel the room acoustic on performed-live tracks, if no-one we've said that to ever understands. The dirtiest muddiest track opened out to reveal it's the hifi being mediocre that can't resolve it, after all your Dream Amp can resolve it. On first hearing an amp like this, to think it sounds dull for it's accurate neutrality is possible, but it's an illusion. Because that's you more used to bright amps, a few minutes listen reveals crisp treble played without the fizz & grain that 'hifi' usually is. We first heard this sound, if with not quite the fidelity, in the much upgraded Trio WX-400U. Unlikely to be found too often in silicon transistor amps for the amount of transistors, if some of the better early ones 1965-67 come near especially if Germaniums. This may sound like some warm Sunday Night mellow out session up in the clouds banging out some Hot Ska Tunes, but it is based on reality: This sound can exist. You can understand "Why They Don't Want You To Have That" because you'd play it at Wembley Stadium levels in perpetual audio delerium, with a permanent inane 'Passed Judgement Day' grin. No, misery & conforming is for you, my lad as is inaccurate audio screetched out by cheapness masquerading as five starrery by the shysters of the audio press. In terms of rating our Amps as "Excellent" this is so far ahead of that it would just sound like a hack reviewer with a Thesaurus. On speakers, the sound is diffused by the room as always, but the huge dynamics will frighten the life out of the cat, so best keep a few spare. The sound instead of being from a vertical wall of speaker, is now on a horizontal 3D plane with the sound laid out in front of you with the sound now disassociated from any sense of origin as the speakers themselves & just there but not connected. The sound is loud but not there so you can't hear it is anything but what it plays, a little eerie. As you can see, the deep revealing sound of valves is a little hypnotic.
Valves Or Transistors... ↑
The eternal debate. Our opinion is either done 'properly' can be as good as the other. Valves are more for the 'thinking' hifi buyer. Transistors are easy plug & play and range from better than valves to truly awful sounding. Valves get hot & generate a lot of heat in a room in the Summer. Transistors correctly adjusted will barely get warm. A Valve amp has a simpler circuit so less to lose the 'magic' of the original recording. Transistor amps are usually overdesigned to get unrealistic specs that lose a lot of the 'magic' of the sound into a dull flat boring sound. Build two amplifiers with the same amount of amplification stages, one Valve and one Transistor & get it wrong picking them out in a Blind Test. They will sound identical if correctly designed, if the educated listener can still tell them apart. We've hear boring sounding Valve Amps; We've heard rough sounding Valve Amps; We've heard grainy sounding Valve amps: it's all down to design & spec used. Valves need regular maintenance with biasing & replacing. Transistors just sit there for decades. Valves look pretty and are on show. Transistors are hidden away, no amp displays it's transistors. Transistors are capable of delivering higher power & more current, there are 200w transistors. Valves generally are capable of 25w pp EL34 & 40w pp KT88 though much higher power ones exist outside of domestic hifi. Transistors can be Germanium that age badly. Valves often get the lettering worn off. Transistors need high amounts of NFB to sound low-noise, a transistor output stage with no NFB sounds like Niagra Falls. Valves can get away with no NFB & still be almost silent. Valves soft-clip better in terms of distorting. Transistor clipping sounds harsh. Design an amplifier with the same amount of amplification stages as Valve or Transistor & get it wrong picking them in a Blind Test. Valves have a finite current power level. Transistors can vary allowing further design ideas to come alive that Valves can;t do. Nobody is going to design a Valve Power Amp with over 20 pointless Valve stages in it like some overdesigned Transistor Amps have. A valve Power Amp gets by with a driver, a splitter & 2 push pull, less is more. Valves need close matching & biasing to sound their best. Transistors you can just throw in & they work right even if their measured gain (on NPN-PNP pairs) is hugely different. Transistor Amps that are (Semi) Complimentary coupled directly to the speaker are easy to damage by shorting outputs, you have one second's grace, we know. Valve amps have a transformer output & early Transistor ones have Capacitor coupled outputs, but we're not shorting them to see what happens. The best Valve Amps are Ultra-Linear designs. The best Transistor Amps are ones without Differentials, if the differential isn't the weakness as the Teac AG-6000 proved. People will be happy with a lousy valve amp as they've been fooled into thinking weak design is 'better'. People will not tolerate a lousy transistor amp. Bad distortion in Valve amps still is listenable. Rough grainy distortion in Transistor amps is unlistenable. A Phono stage is better with Valves than Transistors. Valve Bass is said to be better than Transistor Bass. Transistor Bass can be as good as Valve Bass if the circuit is designed right. Valve Treble is clean, sweet & detailed. Transistor Treble can be clean, sweet, detailed & more powerful if designed right. A Valve stage sounds more "real" as there are less components than the majority of amps made. A Transistor amp is usually poorly designed hiding the fact that a Transistor is actually more versatile than a Valve, there, we've said it. Valve Amps appear loud for their power rating as Transistor ones have higher NFB & more circuitry that actually loses the volume. But on loudspeakers, two 18w amps, one valve, one transistor, only really gave the same value with driving a speaker, on headphones it was an easier load for the valve amp so it sounded louder. Which one you prefer depends on what you've read in the past with opinions formed hard to break. We like valves a lot but are finding the best transistor amps, upgraded naturally, are now matching or even surpassing the valve, but then the valves get another look to better. We've spend ages upgrading valve amps from various years 1963-1993 and in general, the transistor amp delivers the best sound the easiest way. Valves are more for eccentric tinkerers.
*NEW* Transistors or Valves-Tubes? ↑
A LATER OPINION IS...We upgrade all the Amps we get now, reason why is often beyond making a better amp to sell. It's Our Research & some Upgraded amps we sell are these 'experimental amps', these are sold as why keep every one as we are dealers too. With some of the better amps we've had & that runs the full year range 1963-84 it is possible to match the sweet detailed quality of Valves. The area which Transistors will better valves is Bass. Valves such as ECC83 & EL34 are designs from the 1950s-1960s & the limit with valves is current, mA if the plate-anode voltages are high to give their wattage. Transistors can be made much higher current with a lot of 'headroom' to deal with peaks, as long as the power supply is of very high spec. In Valves we can get into clipping surprisingly easy, even if the stage is designed to be clean at 0dB limit, clearly 0dB is not a limit as it is with Digital, so Valves & Transistors can overload & this behaviour depends on the spec of all in the amp. To get the sort of dep full bass that some late 1960s amps can we in theory could do with 'circuit manipulation' such as the Tivoli radios alter the EQ to sound unlike what you'd expect, but it's false. Designed cleanly, the Transistor just has more Reserve to deliver effortless bass. We do try some Transistor amps on our Speakers for daily TV use so
are very familiar with the sound of some. Using our 100w custom-built based on Tube Technology power amps, the sound is very clean using another amp as a preamp, but even at 100w the bass is no fuller. But you'll have heard of Valve Bass being fuller? This is actually playing aged low spec amps & the flabby bass is due to running out of power, the low spec of the design as well as valves past their best that lack treble so will sound bassier. A valve amp done right will be as clean as the best Transistor amp, but Bass is better with Transistors in driving big speakers as this takes current. Valves do have a full bass but it's a lot tighter sounding without the freer current & we've certainly tried in designs. The best Transistor Amps for Current are those with more than one per pair, as we've noticed the 1965 50w Sony TA-1120 with this does sound a lot more controlled than a similar 1968 40w Dokorder 8060 with the standard one pair per channel. The fidelity is similar, the "kick" the Sony delivers does better & having tried others with doubled output transistors like the 1975 Pioneer SA-9100 after much upgrading, that 'extra kick' to drive big speakers is noticeable. Does this mean we recommend 200w-500w amps? No, why do you need 200w, in normal use you'll barely use 2w & only pushing it hard you may use 50w, though speaker sensitivities do affect volume more than wattage. We've not been too keen on 100w+ amps as these rely on too much circuitry.
*NEW* Valve-Tube Configurations ↑
Looking further into Valve amps reveals Guitar Amps are far more advanced. One time where Forums are useful, simply as Guitar Amp guys use their amps & there are ways to tune the sound. Most Hifi Valve amps use an Ultra-Linear output transformer which gives another NFB stage to the amp & all Valve amps have the typical CR one. It sounds very clean but doesn't sound as open as the Trio WX400U does, overall not as musical. Can you undo UL to "old style"? Rogers 15w HG88 Mk III is UL & we thought it wasn't too strong on bass either. The Rogers 8w Cadet III isn't UL & we remember having fun with one of those before. The difference is instead of UL taps, the HT is dropped between the Transformer voltage input & one of the Grids & the Trio WX-400U has a similiar design & was a bassier sound, not one to just mess around with as 400v+ so we'll not give detail beyond what one aware could discover. Other Pentode designs differ as does the simpler Triode version, using EL34s. Triode mode is possible too but it has problems "compressed & midrangey" which is how UL sounds to us knowing the differences between LX33 & WX400U. Pentode, UL & Triode. Pure Pentode without UL sounds better to us & Guitar amps having different needs to Hifi do Pentodes without NFB to get that huge volume. You can go between Class AB (normal use) & Class A but the heat Class A valves put out was excessive said one customer of ours who bought one & swiftly sold it on, even in non-Summer weather. Transformers will differ if in reality so why you need to go Class A even with Transistors as per the Yamaha transistor amps, it makes no difference once the amp is properly upgraded. As for Ultralinear, Wikipedia describes it well, it gives a lower distortion as the extra NFB would, but we've noticed it's at the expense of sound comparing the LX33 & WX400U So we'll try it...
*NEW* ULTRA LINEAR vs PENTODE ON EL34 VALVES. ↑
This is the comparison. We are familar with the Non Ultra Linear sound from the Trio WX400U & others, to swap configurations isn't too tricky. Ultra Linear involves having NFB from the Anode to the Grid, but Hifi Amps already have MFB from after the transformer to an earlier stage, so why add two lots of NFB? The reason why is it's cheaper to get good specs with using low spec parts. Be aware pre 1970 Valve amps don't use Ultra Linear & some comment on them sounding "flabby" on the bass as well as being warm. The WX400U done correctly was far from that. if it was hopeless on original spec unless you heavily limited deep bass. Doing one Channel first to compare quickly, Pentode setting is louder, but to do both channels. The sound is VERY different & Bias only needs the tiniest of adjustments, so you could fit a switch for Pentode or UL mode. The sound is much more natural & "easy", like undoing a tight belt after your Curry, things ease up down below... Here the sound gets the same treatment, the over-compressed "flat" sound is mellowed out into a far bassier sound with treble much crisper, the "hardness" of the midrange. That's playing Reggae with the Bass far more to how the best Transistors can sound. Onto Rock in Stereo, the sound again hugely different, the too upfront Rock sound that many will like is gone, leaving again a far more open sound. Stereo is much wider in Pentode, the sound just has much deeper resolution. It's deceptively subtle as dynamics hit much harder now the unneccessary NFB of UL is removed. The Music sounds Exciting; The Music isn't compressed. Some tracks sound very unusual after hearing them less open sounding. Now having adjusted & played for a few hours, the sound is so different, huge but 'real', Rock Guitar has full weight instead of the scratchy noise it's now without the rough harmonics which can deceptively leave it sounding a bit dull, but then real crisp treble proves it's accuracy. Guitar Amps don't use UL if they use overdrive & other non HI-Fi techniques. The glass valve cases appear hotter than before if no unwanted glowing. We still need to test it further to be sure any distortion is the recording not the response. The sound of this is one that will annoy the neighbours even at lowish volume as it's so huge on the dynamics & the bass isn't limited. One thing we noticed with our upgrades that there was instability in UL that isn't there in Pentode mode. Vinyl can put out subsonic bass doown to 4Hz regions & valve transformers usually aren't much good below 20Hz & once the amp actually went silent on one channel if it returned unaffected after switching off & back on. This instability is related to feedback where too much NFB can be a problem, see elsewhere for the tech reasons. Very likely the power rating is a bit higher too as UL does limit power as the lower volume reveals. Before readers go ripping their UL valve amps apart, the Forums reveal the idea is Non-UL is great for Guitar amps & UL is best for Audio. The reason here is the spec of their Valve Hifi amp isn't great or designed very cleanly on the Preamp & Driver stages, so here is where the UL covers poor design as NFB can give great specs if mediocre sound as 1980s Transistor or IC amps prove. Our 100w TT Monoblocs we always thought lacked the punchy sound of amps like Sony TA-1120 (1965) & the Pioneer SA-9500, the Non UL Pentode mode will get tried on those too. Are we going to fit an UL switch or use UL again? No... UL is flat & lifeless, has no bass & is hard sounding on the speakers. The circuit will probably need a little more tweaking, but UL is gone forever & only good to "better" the spec of mediocre designs. The Luxman LX33 despite the wordy hype on the fascia is a very lousy design as original.
The Best Sounding Valve Amp ↑
What do you get for your Money getting something beyond a starter-price amp, like the ones you see under £1000 on ebay? What it will sound like is like the sound has bypassed your ears & goes straight into your mind. So solid is the sound you could touch it. Nothing unresolved, nothing weak, nothing grainy. Bass fuller than any Transistor amp simply as minimal amplification stages & a low NFB. A valve amp can get away with no NFB beyond the main one. Sadly none of this you'll find as shop-bought as designs these days are lazy & boring, still taking the 1950s Mullard design & still thinking low spec & being afraid of Bass is what Hifi is. Only upgrading will bring out the best of a valve amp & sadly it'll bring out hum & hiss that the weak design spec hides. Every trick needs to be thrown at it, no compromise, no spoilers. Then it'll sound like it ought to. If that sounds far off & scary, then that's because it is. This sort of sound is possible to get in a Transistor Amp to a "not far off" degree, but the thing is you need to understand this sound before you can recognise it. For the decades Hifi mags used to say Going To a Live Concert is how you experience the sound of real music. No it isn't. You are usually a long way away from the music & the room resonances & damping by hundreds of people means you are not hearing music correctly. Only by being in a Studio when the Music is recorded will you have an idea of how it sounds, only an idea mind, as music is recorded with more than one microphone & mixed, you are a sole being standing in one place. This gives the idea that Sound Quality depends on what the person knows or cares about & is entirely subjective based on how good their ears & mind are. With Uneducated Ears & Woolly Brain this is why few really understand Sound Quality...
Class A Valve Amps? ↑
We heard of a buyer who got an Audio Research S6, the later one with the row of valves at the sides, not the front. It's 30w Class A using 3x EL34 per channel. Underneath it looks very modest for the high RRP of nearly £3k. We didn't get to hear it, but the owner has had some of our amps now. In the Yamaha Class A as original, a difference can be heard, but once upgraded there is no difference & we've listened hard. Fiirstly the Class A as with the Yamaha CA-1010 type amplifier runs very hot when the Normal setting is cool. Valves chuck out enough heat as it is with the valve heaters, but with Class A the valves are always 'on' as Class A dictates. But the Class A EL34s even in cold weather chuck out far too much heat to be used in a domestic setting. This particular amp was sold on after a few months as it became tiresome for the heat, no controls such as Tone or Mono and no Phono stage. These minimalist amps are a power amp with a volume control & a few line souce inputs with the selector. But music isn't always perfectly balanced & neither are rooms. The huge 25kg weight of it doesn't help either. Looking at the photos to see what is upgradeable is difficult, the very small capacitors will not give a strong sound, the lack of power & controls was the issue to sell it. This is the trouble with modern valve amps, they are heavily flawed beyond sounding interesting & different but to live with one is clearly difficult. The Prima Luma Prologue 2 we had to upgrade & with some changes it did improve but overall the circuitry was designed by someone not aware of how sweet & effortless the Best Hifi can be, yet it can still deliver power & a good punchy sound. The PLP was driven too hard to get the volume so sounded hard, we could have improved it but at the risk of bringing up other weaknesses. Valves are very difficult to get right. The Vintage ones are too old with very aged spec & design, we upgraded the Trio WX400U as elsewhere on this page but were still not satisfied with it as it needed more gain on Aux level. As with any Hifi, people are looking for their perfection, but generally you'll never find it unless you get someone to upgrade it for you, oh dear we ended on an advert for us to upgrade your amp, but realistically, there is no other option than to get something good & have it made better. But then never really be able to judge the outcome as upgrading often brings out weaknesses that are difficult to overcome & it needs taming back a little.
Valve Amps Need Earthing. ↑
We've noticed quite a few Valve & transistor amps only have 2 core mains, usually a double cable with no markings. The accepted idea is if you can be sure the amp gets earthed elsewhere then you don't need to earth every item. We're making no suggestions of what to Earth or not though as it can compromise Safety. The safest option is Earth all items but use only one main socket for all to avoid hum loops, eg all on one 5 way block. The Trio has the USA type black cable & is usually Earthed via the Soundcard. But using it twice now on the TV it goes bang. An ozone-mains smell is noticeable if on inspection there is no issue or damage. The HT sparking in the valve sockets makes that smell & the Rogers Cadet III 2-parter had that same smell if it was properly earthed. The issue is the TV & TiVo aren't grounded as the Cable input has an isolator to stop a hum loop so it relies on an Earth connection. The Trio needs 3 core mains cable too as hearing it go bang connected to speakers is very scary even for us. The only issue is those odd push on clips called Strain Relief Grommets are now hard to find & though you can reuse the old ones, care is needed to use wire thin enough if not too thin the clip doesn't work. UK standard is for 3-core Mains though plenty of post 1980 item10s are Double Insulated & use 2 core mains. Read more online including the IEEE Safety Codes.
Valve Power Ratings ↑
Transistor amps have high ratings usually because they offer a heavy current delivery in late 1970s days when low sensitivity speakers were popular, an awful idea says we, and a 300w amp was needed to drive the stubborn things. Ass kicking in more ways than one. 300w will be no louder than a 100w amp but be able to deliver a high current. Big deal. A valve amp runs through a transformer, so there can never be any DC on the speakers as the transformer only lets through AC which is sound, though heavy clipping can give near DC & still fry your tweeters. The valve amp doesn't deliver high current so it's modest power ratings are not related to how a transistor amp plays volume wise. From the amps we've had, 10w is the equivalent of a 35w transistor amp, ie a Leak Delta 70. A 15w amp is about a 50w transistor amp & a 20w one is about 75w. This can be found out by playing enough amps to know what volume they can put out without clipping or getting to sound rough. Many transistor amps don't play as loud as they should, see the Power Ratings page for more.
Valve Hiss & Hum Background Noise ↑
For vintage Valve amps, the -db noise levels can vary hugely. A modern hifi item can boast -100 to -120db noise levels so you hear absolutely nothing via speakers of 92db or less, as the maths would suggest & even loud PA speakers with 105db can have a silent background if items are chosen. The Sansui 3000A is noisier as it's a 1967 with -75db on Aux & -70db on Phono, though upgraded the noise level is nearer to -85db at an estimate. A 1986 Yamaha A-720 is -106db on Aux & -92db on MM Phono. We can't find db ratings on valve amps though 1961 Trio KW-60 one states it in mv with Aux as 10mv noise & Phono-Tape Head as 80mv noise. The thing with vintage valve amps is they can be run on 'cool' spec to give very little noise, but to optimal spec the background noise can be quite high to perhaps -60db, but when music plays you'd never hear the noise but for music with lots of 'quiet' the noise will be obtrusive, assuming correct Biasing having being already done. Valve Phono stages can be designed to give very low noise levels though for a MC Phono the level a MC cartridge puts out can be a problem, though we've never heard of any hair-shirt wearers using a MC cartridge to try one. There are various methods to improve the background hiss & hum but with a vintage amp generally being 10w to 20w to run it 'hot' to get a decent volume does bring up the noise levels. On ones we've tried you can make the background almost silent but lose so much of the volume & drive. Yet with a modern 100w amp you will have a much better -db rating if it'll never be as silent as the heavy NFB transistor amps.
Vintage Valve Amps are Sadly Usually Not Worth Fully Upgrading Unless It's For Your Own Use. ↑
The dilemma with pre 1969 Valve Amps is basically they are really not good enough to just lightly recap or upgrade. To recap them to keep them going is probably the best idea. To fully upgrade is a total rebult, more or less a strip down to the sockets & tag strips & redo it. The Trio WX-400U we done all we wanted to over a few years & gave up three times before deciding to try again & as we got it back we've upgraded it even further. We have done a huge amount to it & for the work involved no-one would see the upgrade cost as worthwhile as it is huge. But we try these things for our research & once we feel we've gone far enough without going into more noticeable alterations, we see job as done and it was excellent but knowing better sound elsewhere it was still lacking until the very latest upgrades. We've been looking at other vintage Valve Amps & Receivers and after the disappointment of the Sansui 500A when the later 1967-69 Transistor receivers are great as well as considering several other Receivers including Pioneer ER-420 the idea is beyond the Trio WX-400U & some of the other Trio ones, these just are not very good. Poor Tone controls, AC heaters, ugly looks, bad connectors as well as other people's fiddling makes the odds we'll get another one very unlikely. Plenty of ones 10w-15w but with the Trio an 18w there is little point as we know the 18w Trio doesn't play loud enough for use on the Tannoy Golds & we can hear it flatten out on headphones. For lower levels the Trio sounds great though at least 25w is required for using louder. There are very few Integrated Amps & Receivers higher than 15w as our Amplifiers & Receivers listing pages show. Any amp with a Valve rectifier cannot be upgraded to our high standard so those made 1963-67 are the only option, unless the job of taking out the Valve Rectifier & having higher voltage issues. By 1963-64 Radford are making 30w-60w amps though they'll be the with-preamp type & the early preamps are awful. Beyond PA & Pro amps only really the USA ones like Fisher, Scott & Sherwood as well as the Trio & Sansui stand out as anything better. Many of these never made it to the UK with loads of Trio models & The Sansui 1000/1000A initially being the best, until finding how poor the 500A was, beyond the Fisher & also McIntosh ones. So that's about all there is beyond the UK made gear that is just pre-power units for building into cabinets & the trouble with most of those is they are too limited to really upgrade into something special & too much away from the original design would be needed. This is why we keep rebuilding our 1993 Tube Technology set up as there really isn't much else in valves, to get a useable basic amp & rebuild it to your needs, not the safe lazy spec TT put in theirs.
We see good prices being paid for 50 year old Pioneer & National Panasonic valve receivers, ones in nice external condition on ebay late August 2015, but you can't get past the age & low spec. Pioneer SX-800 & NatPan SA-51. As one who has rebuilt a few amps, to buy one & upgrade it to sell is far too much work & it'd price it out of the market which is still very young on proper quality upgrades. These sort of amps are usually bought by those who write about it on Forums, some doing a good job but more often you wish they'd not mess with things this advanced. Even the 1979 Luxman LX33 needs a total rebuild as it's not as good as the hype on the front suggests. To tackle one of those early FM Multiplex ones from 1963-67 is a huge job but the results are worth it if your skills are good. But if you're not an advanced tech, please leave them alone! Expect to take months to perfect it even if you 'know all the ideas' as no two amps are the same & recapping those double & triple capacitors is mindbending to do it properly & neatly. we had considered getting one of the NatPan ones before but seing the circuit, to us it's the same-old low spec & full rebuild needed which doesn't malke good economic sense, if us rebuilding a Rogers HG88 Mk III is worthwhile as we've worked it out before. The new & unknown amps take a lot. Even the Trio W41, the amplifier version of the Trio WX400U was not as simple as the amp is tiny & doing it right was not easy & all these early valve amps have those useless 'conical' phono sockets that break modern able plugs, so to redo those is a headache too. Read this page & if you think you have the skill & insanity to try one, put yourself to good use & try one of these early Valve receivers or amps as the results if done right will certainly please. These valve receivers make good dinner plate warmers too, as the top lid can reach 60°C which will keep your dinner warm in winter.
Valve Amp Problems In Use ↑
Assuming you find your Best Valve Amp, living with it isn't easy. A vintage one will often smell as the grease paper constructed old transformers get warm as do the remaining older parts. Then valves need occasional Biasing as the bigger output valves start to lose emissions but can still be good once rebiased. Then the problem of some valves like EL34s having poor solder connections from valve wire to the fitted base so the heaters go o/c as they draw higher current. If the pins are that poor it could risk more problems. Then the issue of heat. Nice to warm the room in the Winter, but in Summer it gets a bit much, our TT amps have 8x EL34s & 8x ECC83 that chuck out a lot of heat. Buying new valves can be expensive even on those ones but if your amp uses obsolete ones or expensive 300B KT66 KT88 etc there are equivalents of the KT66 but others cost a fortune. Small preamp valves can last for decades but in regular use the EL34 lasted about 2 years before going too low emission to get the bias. Problems in use with valve amps, our much rebuilt TT 100w monoblocks we purposely done as visibly hardwired on the underside to be sure no breaks as PCBs & valves aren't so reliable for the voltages. But ours was on for 5 minutes ready for TV watching with no sound yet & a worrying 'thwoop" noise got the attention. On checking it expecting issues, there was nothing at all wrong, no duff valves or anything. We put a fuse on the HT & it blew that harshly, but now it works again as we type. No loose wires & the other one was still working so not a mains issue. Over a week later it was still fine, so why did it fail?
Valve Amps and Headphones Woes. ↑
Using Headphones with Valve Amps can cause problems. Many valve amps require a Speaker or Headphone load to work right. Amps with a dedicated Headphone socket like the 1979 Luxman LX33 are difficult too. Adding a headphone socket to a Rogets Cadet III as a few we've seen have had done, unless done right can cause issues. The LX33 switches speakers off with headphones used but on testing the output via test tones on speakers with no headphones to find out why it was dull sounding it reveals the issue explained below. Very few amps have a fitted load resistor to overcome this problem, the Trio WX400U & W41 do. some of the early transistor amps have this loading too, the 1965 Sony TA-1120 & 1967 Pioneer SX1500TF do if usually transistor amps don't need it if it's not in the design.
This is us working this out for ourselves, we'd not expect later amps to be designed with no safety load, after all Trio worked this out by 1963. We got tired of the 1967 Sansui 500A trashing the output valves, but now we know why. It appears the 500A, and other valve amps, have a strange switching circuit for when using headphones & setting it off speaker mode it makes the headphones too quiet, though what are 1967 headphones like? For one who is used to how Transistor amps do headphones, why would we even forsee a problem, until finding out the hard way. This switched setting on the 500A puts an 8 ohm internal resistor load across the output, the tall white thing inside. Without it, the failure is the HT current on the plate rises, so trashes it & the 500A is very high HT for the valves spec already. Obvious now. "You must always use a load with tube amps because, without it, the primary impedance will approach infinite resulting in an excessive voltage on the plates of the power tubes." says one site. Again, that's a catch-all comment to cover design before headphones were popular by the transistor era, no other valve amp we've had has ever needed loading & even our TT amps read steady using the headphone box. The 1963 Trio we see has a set resistor in the design, seeing it was part of the circuit but as it was there thought nothing of it or felt the need to remove it as it was clearly to load the circuit, to therefore make it not need this clumsy switched resistor loading. The Pioneer ER-420 is similar though. Trio got it right therefore as there is no need to load the amp as their design allows some impedance to be always there & not affect the volume noticeably & we're not trying it without any load. Interestingly, the Trio WX-400U manual doesn't appear aware of this, or it was added into later versions only as it warns "Never operate the Receiver without first connecting a speaker to each channel output" and for Headphones "Never leave the Speaker switch in the Off position if Headphones are not connected or in the On position if speakers are not connected. This may cause damage to the output circuitry". But the later versions wisely add the load resistor & looking at several amp circuits, it's the only one to use this, mainly because the designs are too early for headphones, the 1963 Trio appears to be one of the very first, ours has 1964 dated pots. All other valve amps just have speaker outputs with no loading or warnings. All the Rogers HG88 Mk III booklet says is an order of connecting involved speakers before mains, if no warnings. Looking at other Trio valve amps the story differs between two versions: a switched in resistor or a fixed one. The fixed one clearly the best idea to stop errors. Again, no-one tells you these things or explains & us piecing together bits of what we discover & the penny dropped with the first quote. To do it the Trio way to be safe appears the best solution, again the Trio WX-400U we have is once again a winner. In comparison the transistor Sansui 3000A uses no load resistor though with 2 sets of speakers it uses an idea beyond this section, though it needs a speaker load to read DC offset correctly. As noted elsewhere the Sony TA-1120A has a poor headphone socket from the preamp & we wired our one to run from the usual way with no problems. Just early design progressions. The transistor 1967 Pioneer SX-1500TF uses loading as does the 1968-70 Sony STR-6120 & the 1969 Sansui 4000. The Sansui AU-999, Teac AG-7000, Trio-Kenwood TK-140E & KA-6000 don't.
Valves: When To Replace Them ↑
If you've got an amp of 2 years or more in age then there is a chance the valves, especially output ones, could need replacing. When we used to keep ours on all day they lasted about 2 years before the EL34 Bias couldn't be set correctly as it was too weak. In those days, sellers didn't run them in & grade them & failures as well as ones reading too high bias or even glowing far too bright on the grid were common. Buying any used Valve amp to replace the valves as yours are 'unknowns' sadly is the deal, so factor that into your buying cash. The smaller preamp valves can last much longer though we used 1960s Yellow Mullard ECCs that are now sounding soft compared to brand new ones. How these actually test on an AVO meter or similar, we've never tried as noted elsewhere on the page. To see if the dull sounding Mullards would still be Good or whatever isn't worth paying £££ for a meter when a set of valves isn't too much. In one amp we have as of typing, we've altered it a bit so it needed 3 new power amp first stage ECCs, but the preamp with the original valves still compares adequately, but they'll likely still get replaced to keep all the same brand. The EL34s are late 1970s Matsushita ones 6CA7/EL34 marked. Having upgraded so far, the amp was sounding slow on treble with a thich bass that wasn't too smooth at times. Putting in another set of EL34s from the TT monoblocs, these are old now but bias good, the sweet sound returned with a well balanced sound. Going back to the Mats. ones the poor sound again. These must be at the end of their life as Biasing is up high to get almost the right spec, so they start to lose fidelity as well as the treble & bass extremes as they start to go weak. You really don't need a meter to tell you the sound is not it's best, but having spare valves to compare is your cheapest way. We'll be buying EL34s, not E34L or some other 'tuned' design that may please Guitar users, but we want the Neutral sound based on the older designs.
Valve Rolling within ECC81, ECC82 & ECC83 ↑
The try it and see highly risky art of swapping similar pin valves to see simply which sounds better in a particular stage in an amplifier. Generally ECC83 is the one most used but ECC81 & ECC82 are useful & quite different. For our rebuilding of our TT power amps we tried all three & found for a power amp the ECC82 works best for the input & splitter valves. This was confirmed by using all three in the Trio WX-400U & finding the ECC82 has more gain & solidness to the sound than the ECC83. The Trio had ECC81 used briefly as it's Tone stage valve but was found to lack drive. ECC83 went back but a new Sovtek sounded a bit rough. an old Tungsram one from a Heathkit from years ago was much smoother, though that newish Sovtek was found weak used elsewhere. Oddly our yellow Mullard ECC81s were a bit weaker on comparing than the Edicron ones that Maplin used to sell before they went too High Street. So the TT pre & Trio now have ECC82s that give a louder output that is more solid so has better drive. The specs of the 81, 82 & 83 differ with the ECC81 a medium to high gain at 60μ RF mixer valve with a 10k ohm anode load quoted. ECC82 a low gain 17μ audio stage valve with a 7.7k ohm anode load though using a higher resistance can alter it's characteristics. ECC83 a high gain 100μ high quality audio valve with 62k ohm anode load. These specs can be found on datasheets together with suggested circuits. But the mysterious factor of impedance of the stages before & after explains why a low gain ECC82 sounds louder than the ECC83 as well as current capabilities. But after some listening with the Trio the ECC82s apparently lacked drive even if they were louder. ECC83s back in & the sound was thinner. ECC82s again & turned up louder there's the drive after all as the extra volume is filling in the ECC83 thin sound. The ECC82 has proved a better match. Valve rolling is as subjective as any amp comparing & easy to get things wrong.
Valve Rolling with different number valves ↑
This is where the Danger & Trouble can start. Vintage ECC81-82-83s are much wanted by some users, if we've found the modern JJ ones to better old Mullards for better sound smoothness & dynamics of sound. But there are a lot of "Equivalents" left sitting around unwanted & some are said to be Military Spec or Better Spec, but according to who? They will Sound Different, whether any will sound better is Questionable as different specs as with any Amplifier means it will sound different & in our listening-comparing sessions it is so easy to fool yourself that one is better until trying others another day & finding "the best one" was a false prophet & it sounds harsh in comparison to known references. The harsh amp's sound will have got used to & then the references can sound wrong as your hearing auto-compensates. Onky with several compare sessions over several days can you really be sure of the best ones, as in amp testing & tube rolling. Personally beyond trying the mix of three ECCs above, we don't tube-roll, we alter the circuit instead. Also when tube rolling, you need to Rebias the output valves & the preamp ones too if possible. If not, you'll either undercurrent or overcurrent which can lead to ruination & severe overheating. The Trio WX400U we got back after the owner tube rolled & half trashed it. So a survey of the valves he used is interesting. First is a Sylvania 5814A, said to be a 12AU7/ECC82 equivalent, well it's a different valve as it's not marked ECC82 & it will have it's own circuit requirement, a bit simplistic, but we don't have a stash of NOS ones to sell. This draws slightly more heater current, as per a TubeWorld.com site giving honest differences, so we'll use their info save raking thru spec sheets. So the 5814A will upset power supplies, in this case the WX400U was designed tight by us to the available current & this will have upset the circuit balance as the DC bias circuit was run from the same source. This will have dragged down the Bias voltage which is as bad as pushing Bias too hard & the EL84s glowed medium red & scorched the red valve print to brown. 5814A not recommended therefore as it's too far off spec. The next ones were Mazda 12AX7A also labelled ECC83, the added 'A' means a lower noise version, so this one is good to use. Then the Mullard M8137-CV4004 KB/DC supposedly an ECC83 with Military markings but only by sellers having boxes of them, these are offered at around £30 each but no spec findable, but they aren't marked ECC83 so could be anything so no good to try & risk, if Military Spec Valves & Transistors do exist, to check the spec sheets to be safe rather than just put one in as "the seller says so".
Valves can forever be Tuned. ↑
Guitar amp users play on this characteristic to give varying tones from a Valve. On spec it's right for Hifi, under spec it's warmer & softer & over spec it's rougher & edgier. Running Hot & Overdrive are the terms used. For Hifi to spec is the best way, if altering amps like the Trio WX-400U you find you can upset the balance very easily in altering a circuit but there are loading methods to alter this. Putting in different valves to the original obsolete ones is a very difficult job as the transformer & power supply might be inadequate. Enough of a clue there to deal with it & go find out more elsewhere, but we don't say too much of the tech stuff as we see the problems Forum tweakers get into. The Spec sheets are your biggest help in optimising a valve amp as is a multimeter. Do it on all stages to give it the 'sound' you want. Transistors equally are Tuneable also in a very similar way which is why some amps sound so different to others, to tune every Transistor is possible if a huge job.
Valves: Adjustable or Fixed Bias? ↑
The output valve requires a set amount of bias to work correctly for the given output Valve. Care needs to be taken not to upset the current-voltage balance by careless Tube Rolling as some think is better than the valves the amp was designed for, see more elsewhere on this page. So you have a 1960s Valve-Tube Amp with Fixed Bias, it's to the design so it will do. And it will do OK but it's not as optimised as having both halves of the Push-Pull pair biased exactly as you get with adjust pots reading off the circuit. The Rogers HG88 Mk III has a very crude balancing slider to do similar. Ideally each valve of the pair should be within 2% of the other. You can get closer if the older amps might vary in use. Having known adjustable Bias valve amps, the treble is fast & crisp, in fixed bias it's a softer mellower sound which is touted as the Valve Sound. But it's just not adjusted optimally. So one Fixed Bias 1960s amp, that already has DC heaters from the aged AC & Hum Balance old ideas, got adjustable bias too. The Bias L to R can be within 10% as it doesn't make too much difference, or you can add extra parts, but the P-P pair must be adjusted to as even as possible. The rewards for the effort is well worth it as Fidelity is hugely improved. All this is naturally very advanced work but the improvement is huge: Bass has the deepness it was lacking, Treble is much faster & more punchy as more crisp. To Bias valves is critical, there are Bias Calculators online like this one which is easy enough. Read the Plate Voltage which is the HT voltage from the Output Transformer to the Anode pin, not necessarily the main HT cap voltage, input the valve number & it'll give you a value for Optimum setting. You can run it Hotter or Cooler as the page shows, hotter is a more upfront sound, cooler is a softer sound but the Optimum one makes sense unless your speakers are a bit dull or bright. It is worse to underbias than overbias & issues such as glowing grey inners (anodes) can trash the valve & amp if you leave it adjusted wrong for too long, the amateur valve amp user won't realise as valves get hot anyway. Underbias can cause tube flashes & red anodes as the valve gets much too hot and something will fail or fry the output transformers.
Not many people want Valve amps... ↑
They are for non-conformists & the insane tinkerers of the world only & we aren't many. You see plenty going unsold & forever relisted on ebay. Some for a good reason too. If you know someone with a Valve amp, get them to Demo it with your known Music & hear how different even a Rogers amp can sound. Transistor amps are far more user friendly & the best can be very close to the best in Valves, if upgraded beyond. Valve amps are more for DIY-ers or Collectors, mass market brand Hifi Retailer shops do not stock Valve Amps & one we know that did have one set couldn't shift it in any of their big London shops until one person whose site you are reading now expressed an interest after being unimpressed with the big Transistor amps they had in desperation tried us with those & we took the whole lot. People see our valve amps & wonder what they are, are they just for show? No, we use them every day & this surprises them, what a Kook they must think. One older Reggae guy wondered why they weren't built into a case like old 1960s ones were.
You want to Buy a Vintage Valve Amp? ↑
If you want to start in Valves, there are really only two that are affordable & findable. The Rogers Cadet III is easy to find but avoid the 2 part one as the connecting cable goes bad. Forget the earlier Cadets as too early. Also the Rogers HG88 Mk III that looks like a bigger Cadet III, again avoid the earlier ones. The Quad & Leak valve amps are straightforward enough too though the preamps are a bit lousy. Look for signs of fiddling, bad modifications, lack of wooden outer cases & long stored ones that could be corroded, as with any amp. The Rogers were used for longer than a mid 1970s amp & will have the odd old repair & replaced valves that isn't such a problem. Missing buttons & switches are rare though we've parted ones out & have some selling. Burnt out transformer replacements are unfindable so hear it work or be sure the seller has tried it. If you like the pretty Trio WX-400U as on our Best Looking Amps page, there are other similar ones if less flashy looking. These are much more advanced than the two Rogers simply as you get a Tuner too. Even ignoring the Tuner they are not as neatly made as the Rogers, wires & parts will look like a Rats' nest & not inspire confidence unless you can dig in deep. The idea is a Rogers will generally be OK to use once serviced. Parts may be aged but they will work. Similarly a portable record player with valves is a very simple design & will not take much to get it going, though dried up oil in the turntable mechanism is another thing as it needs totally stripping down. A complicated receiver will be very different as these are usually with far more parts & the odds are any non Rogers, Quad or Leak valve amp you buy will need work done to be at least useable. To recklessly plug in an unknown valve amp untested & unused usually means you'll blow fuses at the least, so don't do it. Take it to a specialist repair shop & have them get it reliable for you. We don't do repairs. If you can't do the work, really don't bother to buy one & risk ruining it or yourself, but look out for sellers who've restored them as then you have a good chance to get one to use, but on ones we've seen they are shy to show their work. Not easy buying one is it? You DIY it & have fun getting it right is the deal.
Vintage Valve Preamps ↑
If you buy a nice Quad II or Lowther, you'll possibly get a good sound out of it with a transistor preamp, or use the pre-out connectors on most amps, beware some don't have the volume control in the preamp stage. Quad IIs are a bit of a pain as they need 400mV whereas most power amps need 100mV or less. This means you'll have a job driving the Quads to their full loudness, you'll hear a gentle pretty sound, but volume will be low. "Yeah, but..." says you. Reminded of this reading one seller's plight. Get the power amp, but the valve preamp. There are modern ones at about £2000 with no tone controls & maybe no Phono stage either. The McIntosh valve pre we tried as on that page, but then problems of level matching again & the McIntosh have a fairly obvious background hiss if not partnered with their own amps. You could buy the original Quad preamp, but it's a clunky horrible thing & it needs something to keep it still as you push those heavy buttons. And suffer loud thumps as you switch. So what did we do? No post 1980 valve preamp will have Tone Controls for a start. We just tried others but in the end designed our current 1 box pre out of parts of the Tube Technology 2 box & added a tone control, not easy. Before that, we used the Rogers Cadet on the intro page as a preamp. Rogers & the Heathkit valve amps have preamp, tone & phono stages, if the power output isn't enough & you don't want to buy bigger more efficient speakers, there is one cheaper option. For how loud the 15w Rogers HG88 goes thru headphones, it makes many 50w (transistor) amps sound feeble.
You want to Buy a Modern Valve Amp? ↑
Modern meaning any amp made from 1979-80 with the Luxman & Radford revived Valve-Tube amps to ones currently available on Hifi dealer sites heading into 5-figures. Plenty of benefits are here but the cost for the better ones starts at the £3000 level for new, if less as used, so don't think some cheap Chinese one on ebay is anything but a taste of valves & will sound boring to the point you give up on valves. Any amp is only as good as it's design & money spent & in valves there is no cheaping out. Stick to currently available valves like EL34 or KT66 derivatives, the 300B Single ended amps are only very low wattage & a different scene entirely with their sweet sound but limited frequency range. The best valves can give a sound that is full, solid, fast, punchy, tangible & with absolutely nothing reduced or dumbed down, so the Bass goes fully deep & is controlled by good design. To hear that sound & you'll not want for any other fuzzy grainy blurry amp that is all of domestically sold Hifi. But don't forget the idea of commercial goods is "They Don't Want You Having The Best Stuff Ever" and even the Best will have cynical limiters as with Transistor amps. It takes a skilled Upgrader to realise the best in any amp & us having done many upgrades are aware there is only so far you can go with most amps without losing the basic originality. The best Hifi brings a sound to Die For, though be aware we've had many OMG sessions with amps but then find other amps 'better' or more truthfully... 'different' which then adds to the learning as well as confusion. As our Deciding Top Amps page states, your Audio Mind does act as an EQ filling in what is lacking until you compare with others. Hifi is fun, isn't it?
The Trouble With Modern Valve Amps ↑
There are many Valve amps, preamps, integrated amps made since the early 1990s that we consider as Modern Valve Amps. Having heard some & researched many more there are obvious problems. A Naff or Silly Brand name is a big hint, we had a Sparkomatic brand radio-cassette in a car years ago & it was laughed at for the naff name. No credibility. Brands of today in valve tend to be Chinese or American with some European designs. They are generally overrated & overpriced, using lazy copies of old 1950s & 1960s designs, some of which we know have very poor design, yes even the Mullard ones. Today's valve amps are lazy designs, low spec, delusions of grandeur & having had some good brand vintage valves & the 1990s Tube Technology, to just about gut their lousy circuits out & rebuild-redesign 'properly' is what we have ended up doing. These modern amps rarely have circuit diagrams findable as it'd reveal they just stole an aged design & didn't bother to think to upspec it. The Prima Luna amp we have on the reviews page tries to better itself with autobias to allow weak design to be less obvious, but it still uses the poor Mullard design & low spec throughout, as well as poor design to get higher volume resulting in a hard sound. We heard the same owner paid £2.5k for a 30w Class A amp but soon tired of it's remarkable heating-the-room properties & soft weak sound. We saw the inside of the amp online & seeing only low spec design, to sound weak & soft it'll be a rubbish design as is always the case and we weren't too surprised to hear it was soon sold on. One Audible Illusions Modulus 3A we were asked about looks hideous: the huge excess of circuitry & voltage regulators show awful design totally alien to how we've designed valve amps so sound so sweet & lively, this 3A preamp will sound so utterly boring it's just not even worth starting on. Some modern valve amps like Audio Note are an insult for the cheap casework, basic looking design but charging £KK for useless-vanity parts like solid silver transformers & ultra exotic capacitors. as with most post 1980 "Hifi" they are desperate to sell you A Dream, but you'll never be sold A Reality, just some average overpriced 'wonder' that you'll get bored of fast. We soon got bored of the TT amps & started pulling them apart to better them.
Are Valve Receivers Really Just A Compromise? ↑
Stereo FM Multiplex Receivers with Valves were made from 1963 to 1966 mostly, with some pre Multiplex receivers with odd ideas like 2 tuner stages on some to get pre-Multiplex Stereo like on one Pioneer SMB 201 that the seller can't find a buyer for. These can look very attractive in the pre plain metal box ones that were around by 1966, such as the 1964 Trio WX-400U with it's Cadillac styling. But then there's the problem of 3 separate units stuffed into one large cabinet: Tuner, Preamp & Power Amp. For the size these 3 items as separate Valve units & even making 4 boxes with monobloc amps, it's all in one box & the tuner by virtue of the amount of Valves needed for MW, FM & Multiplex decoder, the rest will be sized down. A 10w Rogers Cadet III has small output transformers, but has a good volume, but spolied elsewhere by high input resistors. The Trio WX400U we have spent a lot of time on getting the best out of & it appears to deliver a high quality sound, but this is a quality unit, there are cheaper ones that are not worth the huge effort to restore. Half the space is the tuner, the Phono is cramped up on one side & the preamp is mostly at the front, with filter wires going into the Power amp stages. The Power Amp has ample space but the design isn't great unless fully rebuilt with just about everything redesigned, as noted below. We did wonder above if a valve receiver with FM Stereo & Phono was too much in one box. The answer after the Trio WX400U is the Receiver is actually a better item as there is far more space to put in big capacitors & do it properly. The hardwiring technique allows more scope than a circuit board with layers of cable & components that allow more to be fitted in even if some is a bit cramped to work on. The only difficulty is you have to rebuild the Tuner with upgraded parts to do the amp properly, but then you hear how great FM Stereo with all valves is. Actually to upgrade some smaller valve amplifiers seems less appealing as space is so limited, the Rogers Cadet III & HG88 III are looked at below. The answer therefore is "No" in terms of the 1963-67 receivers we've tried, the amp stages are as good or better with extra space inside. But the limit to a 1960s receiver is the age.
Buying Vintage Valve Amps Is Often A Huge Gamble. ↑
Ones we've had & seen have been subject to years of not being used & were in very poor condition in all ways, used more recently by an amateur who has fiddled with it & not done it properly. If you buy one, the amp will probably need serious amounts of work either way if you plan to use it. As with anything, condition, how much used & how well stored in a dry house means a lot. Saw a 1929 Sunburst radio the seller bought new in 1935, the shop apparently told him the older model was a new one, nothing new there, but it was in top grade in & out as he even kept it covered with the oilskin bag when not in use so dust & air kept out. You can find this sort of grade in Hifi to, if not as often as you'd want. When getting old valve amps going again, ordering & upgrading parts to modern spec, not just blindly replacing like for like involves design knowledge & then the work involved doing it. Advanced stuff. A Simple amp like the Rogers Cadet III or HG88 is not so hard to fix, but a receiver or more advanced amplifier will be beyond most people. Remember you are playing with 250v to 500+v here & unless you know what you are doing, leave it & pay extra & try to buy from a seller who has done the work properly & charges a fair price. But there are almost no sellers like that sadly, all valve amps get worked on by amateurs who can be good or can be awful & even more competent ones just idly replace capacitors value for value not taking advantage of better component values of today. If you buy a nice looking amp, get it looked over before using it as looks can be deceiving, eg. capacitors dry out & don't show any clue.
Single Ended Triode (SET) Valves & No-NFB ↑
What does that mean? A Class A amp with just one valve output per channel, compared to two for Push-Pull though you can have 4 valves per channel as Push-Pull with doubled power. A valve flat preamp is a Single Ended Triode also. Depending on design, these SET amps may have no NFB at all, see above. This will leave the audio sounding very sweet which is a huge appeal of these low powered SET amps, but a truth that is perhaps forgotten is you are only hearing a part of the Audio frequency range, typically a no-NFB output stage will be very loud but only on the midrange region with deep bass & high treble being very much less, though not always the same sound. In terms of graphs of output vs frequency the SET response is going to be rolled off at the extremes whilst one with NFB or Push-Pull will be a straight line in theory. Even our Pye G/GR 1932 Radiogram with it's 4w output pentode valve delivers an appealing eerie sound with probably a lot of distortion as all parts are original, but this sound is what an SET amp will sound like even today. But a SET amplifier may sound nice but as it doesn't deliver the full audio range is it really Hifi? You need to choose a bright sounding speaker with low damping to get a better sound it appears, the more damped Tannoy 15" Gold appears not to sound so good with a SET amp, this is why. On building our valve amps, they were left no-NFB just to see what they sounded like. Apart from not having enough gain on an early version, the sound was very different using the Soundcard direct into the Poweramp & via Headphones. Deep bass & Treble was there, so much for limited range though comparing a known great Sansui 5000X transistor amp after playing Valves NFB-free for a while, the Sansui sounded quite limited with a smeary treble. But the ears had got used to the NFB valve sound & later the NFB-free amp on the Tannoys sounded nice but a bit like a Passive preamp sound it lacked a true volume with Tone controls not doing too much to alter the sound. The Sansui 5000X on speakers sounded excellent with a spot-on midrange with much Tone control variance. NFB-free is a sound you can like & think is the better, but essentially it sounds under amplified & there can be a arying lack of extension at the frequency extremes as a no-NFB respoinse curve is much less flat than one that is. Interesting though.
Tired of the thin processed edgy Transistor sound? ↑
That Transistor edgy sound is not really just the transistors, but how the design is made for Domestic use with cost cutting & low spec. PA style transistor design can avoid this sound. The rich intimate "filled out" sound like you hear live at a small Gig is what music sounds like & Valves done well deliver just that & so can Transistors which may surprise. Find what you think is a clean non-edgy sounding Transistor amp, then play a good valve amp, then replay that Transistor one & you'll hear it is edgy after all. It'll generally sound false & over-designed. Very rarely will you be close or better. On ebay there are plenty of cheap Chinese import new Valve amps out there with crazy selling techniques, though what they are like is another thing. If they let you buy & return within 30 days, trying it for £20 postage is well worth it, else how else are you going to get a Valve amp? A Rogers Cadet III? A Heathkit S88? Both are certainly worth a try, but are now 45-50 years old & with high voltages parts are well off spec.
Vintage Valve Amps Need Fully Rebuilding. ↑
We've rebuilt several now & to fully rebuild them really is the only way unless you are happy with the aged components & feeble design that was based on 45-50+ year old ideas & available components. The risk of capacitor failure is just about guaranteed, from the paper in oil couling capacitors to the 400v power supply ones. They fail, they explode & can trash transformers in rare cases. Not that the original designs Mullard suggest for an EL34 are bad, far from it, but just low capacitance due to part sizes & availability & sticking to old ideals is the problem few dare to take no notice of & do it their way. We do & it pays off very well, but you need to know Hifi design to choose carefully. Read below for what we done with our 1963 Valve Stereo FM Receiver Trio WX-400U over a few years including giving up on it several times. Looking at it now, we've rebuilt the Power Supply, recapped & upgraded all capacitors beyond tuner stage coupling, used a better design for the Tone stage as it was weak, improved the Tone, Driver, Splitter & Output valve stages with better design meaning the resistor values are optimised. It can be done subtly, but in reality what remains of the old innards is just the casework, fixings, wires, transformers & the Tuner stage. The 1963 design was very limited & with badly aged coupling capacitors of very low value, it must have sounded a bit limited very quickly by comparing with the 1967 ranges, but to last just a few years use was all it was made for. See a picture of it on the Best Looking Amps page, it looks cool & retro now, but by 1966 it will have looked so old fashioned for the next two decades. We are grateful to those who kept them safely for their wilderness years. Rebuilding takes design knowledge & only the enthusiast will undertake this work, to pay someone to upgrade & rebuild a complex hard wired valve amp would scare you on the cost & as we know, knowing how good the results can be is hard to know, which is why we gave up at least three times before finally getting it right. But if we decide to do another valve receiver, odds are it'd not have the 'luck' of being able to take all the upgrading. But we did finally with the Trio & made it sound wonderful at last. The Rogers Cadet III, Rogers HG88 Mk III & Armstrong 221 valve amps all had limitations: power & spec just too low, bad input resistor cheap out & overall a crappy amp respectively. It's like panning for Gold, you have no idea what the outcome will be until you go all the way with it & time wasted is a high risk. We have learnt much doing our upgrading & we take little notice of how others do things, as in life. If we got one of the £2000 selling Radford, Lowther, Vortexion or Westrex quality valve amps be sure we'd have to nearly gut them out & rebuild which would bring a far better sound, but probably alienate those who buy them all-original as shelf display items. The last two brands are PA gear & be sure after 50 years they are in such rough grade a rebuild is all you could do. But subtly is the only way. Wanna try? So what does the buyer of these do with them? Put them on a shelf & maybe try them once a year? No. You should get them fully rebuilt like they do nice vintage items & Cars on the USA TV restoration shows & give them a new lease of life as they deserve it.
What a Recapped upgrade Valve Amp Sounds Like... ↑
A few Valve amps have got our upgrade treatment now. The Armstrong 221 wasn't worth the effort, the Rogers HG88 MkIII was great but limited by the lazy design with 220K resistors on Line input instead of having a separate valve for the Phono, the Trio WX-400U done hugely better, read more below, but took so many tries & give-ups but it never got bad problems like some can get. Now we've tried a few valves so know what to do. A vintage amp takes redesign to lose the non-single power capacitors. So before as nearly original, if with new coupling caps to the aged values, it sounded warm & bass light, so sweet & fresh but obviously bandwidth limited. No need for it to be like that & done properly you can be forgiven for thinking it has soft treble until you actually get treble in the music and it just gives what the music has, the transistor added treble air is just how transistors are & need a little fine tuning to tidy the excesses. Valves don't need this & with less amplifying stages the real music is closer to the listener than with any transistor amp. The natural fresh sound with no limits beyond the music source proves valves are the winner, but to find a Vintage Valve amp has been tough as most are too old, too limited or with poor circuitry. What does a Valve amp done properly sound like? It sounds Neutral, no artifacts or eek. bits. As free of adding any character of it's own though it will to a tiny degree. Treble is fully extended & resolved & Bass is deep and natural. On playing TV sound through, you are so free of mush & artifact, on first playing it, you feel a bit confused like the sound isn't properly on as all the mush of other amps in that blanket over the speakers is now lifted & lets in the freshness. Only by hearing it will you know. The huge dynamic range without any ringing or limiting is your Reward. The risk of being Hypnotised by Music so natural sounding is your Risk.
The Trouble With Valves Is... ↑
We've spent a few years learning just the Transistor amps for what they are, their design methods & weaknesses & the best that can be got out of them. Our Valve monoblocs have no headphone socket though you can wire the speaker outputs via a resistor to use headphones & we made a box doing just this when we had the Radford HD250 which was an early one with no headphone socket & also the Rogers Cadet III here & wanted to play them with Headphones. So we've only actually played our Valve hifi with Headphones once & probably early-midway into our trying lots of amps game when we had a few Yamahas here. The Rogers HG88 Mk III got upgraded using headphones similarly but was limited by the high input resistor cheapout. So really we only first heard valves quite recently in all these test sessions by getting the Trio WX-400U done right. Until then it was Transistor amps that got the attention. The Trouble With Valves Is... they make most transistor amps sound quite alien after being used to valves. We've been selling off most of our favourite amps as we're not collecting them & even the ones we rated the highest, being used now to valves, they sound very different. A smeary sort of sound, an odd roughness on an amp thought very focussed before. Because we are forever experimenting, we're doing one amp with purely valve amp techniques, now forgetting the transistor ones for a while & see how it does. The Sony TA-1130 is the lucky beast, it's back with us for the third time so it must want to stay. The sound of Valves done properly is just so effortless, so 'not there' no wall or curtain blocking the performance, no grime smeared on the TV screen, it lets you in. It can initially sound like there is no sound at all, as we've found on upgrading our valve amp years back, the 'no sound' is the lack of any artifice or veil over the sound, huge dynamics are possible and can take getting used to. Some things we changed left an OMG for the rest of the day, but the next day the mind was tuned in better to it. We only use that amp for TV daily usually & often you hear sounds on TV but aren't sure they aren't around you, eg alarms, sirens, thuds etc. Realism improving as you better an amp but usually the bettering is only based on what you are used to, to go fully rebuilt upgrade valves can give a huge dynamic you might not have the brain space for it. We remember one relative couldn't take the huge soundstage even at talking volume as it was a bit too real. So we've rebuilt a few valve amps & offer a rebuild upgrade recap service. Whether anyone will want us to rebuild a valve amp is to be seen, beware it'll not be anything but a high price for the work involved. We may buy more Valve amps to try ourselves & then sell on to a lucky buyer. But after having read the above section even, be sure the same one will working on your amp will try their best to get that sound into your amp & freak you out with the beauty of music you've never imagined. Whether your amp is capable is another thing. We never though the Trio below was capable, but it happened. It was the learning curve.
Valves Need DC Heaters ↑
All vintage amps use AC heaters on the critical audio preamp, power amp driver & splitter. The Tuner stage & Output valves are fine with AC but the rest really needs DC heaters. With the low spec of the original design, purposely very light on bass to hide hum, you'll not be bothered by the Heater hum, but once you upgrade & resdesign parts to be higher resolution, up comes the Hum. Then you have to hope for a miracle that the transformer can deliver enough for DC heaters, but this is advanced design & a high risk as there is a chance the transformer is inadequate.
Valves Need Maintenance. ↑
Transistor amps rarely need any maintenance, a good vintage amp can still work fine for years once serviced & adjusted. But Valves need much more attention. The output valves need Biasing adjusting, as we found on our Valve amps on the last service. The Bias on EL34s should be 330mv, it was low at 190-220mv on all. They have aged & this is why rebiasing is needed up to the adjusting limits before you go & revalve as if you tested them on an AVO meter it'd put them in the 'Change' range. Ours are nearly at this stage. The difference in sound biased right is quite noticeable, bass goes deeper & treble is much clearer. We had some big AVO valve testers years ago with the user books, but they looked so ancient inside we didn't think they'd be reliable without servicing & especially calibrating, so just sold them on. Valves can be a pain too with the Heaters going o/c as the pins on EL34s rely on the outer pin being soldered to the wire coming out of the valve & dry joints can occur. At worst a key Valve pin could go o/c & if not caught quickly the output transformers could get damaged. We know... Sometimes a valve may be lit up on the heater so appears good but show no Bias reading leading to finding the failed component, you could still use the amp if it's a quad valve output & not notice. Valves are worth it though if you can do the maintenance. And can take the heat.
Valve Experts? ↑
We read some "Guru" upgraded a valve amp on ebay. You know from that we're not going to be too complimentary. One of the Hifi News advertised 'World Audio Design' ones. We used to read of these as the Vellman Maplins valve amps in the mid 1990s. The opinion we reached from just reading was they were pretty basic & often suffered from hum as does the one here too. Valve amps always have the very slightest hum as overall NFB is much less than transistors, but the fresh open sound more than makes up for a tiny bit of noise. They used exotic components costing much more than the ones you can buy on RS or Farnell that are perfectly good as we use these parts ourselves. But to then read they still use 0.1µf or 0.22µf coupling caps, again overpriced idiot types "Beeswax Paper Foil" was one and nearly £40 each is insane. We've done plenty of valve work as the WX400U proves & we know what we are doing. Never keen on the BS of life & other people's unqualified opinions. Valve amps can cope very well with sub bass down to 20Hz, only a modest Cadet III might benefit from smaller values to avoid instability. They use silver solder & others use silver cable. You can see they put rubber rings on the amps supposedly to limit vibration. What a stupid idea. Put the amp on good isolating feet is all you need, the rings will only make the valves run hotter as these rings insulate the valves from the air. Saw a guy with a Garrard 301 with rubber rings around the platter edge, proved they dampen nothing by taking the platter off & pinging it like a bell. But you can't tell people they are trusting in rubbish & he put them back on next time we saw him. So it asks the question.. why is the person selling the amp? Only for £800 when they list £450 of new parts not including labour costs. They just revalved it weeks ago. Our opinion of 'experts' and 'gurus' is... don't you dare call us that if you consider we may be. 'Expert' to us means King of BS who doesn't know as much as they think they do & calls themselves that term to ego boost. Beware 'experts' on TV shows like 'Pawn Stars' may indeed know their subjects well, but be sure they read up well to be sure they sound knowledgeable as no-one can know minutae about a rare item though often they will understand it once they see it after reading up on it. Reading up means taking another's opinion as fact, it may be from Wikipedia. Other 'expert' restorations of valve gear use the same value capacitors in the big holders with packing to fit them in & you always just see the same useless small values as the 'restorer' has no idea of design. The owner then wonders why it still sounds a bit lousy & sells it on. It takes design knowledge to know how to do this properly & it's very rare we see upgrades done to our standards.
Valve Tuners? ↑
FM stereo by a rebuilt Valve Receiver or Tuner is a delight, but Vintage Tuners like the Trio WX-400U receiver has is going to be with an excess of aged capacitors, the usual low spec as well as ceramics that are not all required. To recap a Tuner is well worthwhile but risky, the muddy almost AM sounding FM comes alive but it takes a lot of changes. To not go too deep into it, ceramics in several early stages rely on them for the reject effect so don't just replace all without trying it each stage to be able to undo a bad move. How far you dare go is the risk of working on Tuners.
The Difference between the Best Valve Sound and Best Transistor Sound. ↑
With amps of valve & transistor design fully recapped & maxed out there is only a minor difference between Valve & Transistor Sound. But the higher you go with Valves, the more the Transistor limitations become audible. Takes a lot of work to reveal this in any amp though. Valves use less amplifying stages so the sound is less "messed with" and blurred by amplifying the sound & extras, so the Damping Factor is lower & the sound is more natural, meaning it is a Richer Bassy sound. We'll not use "Warm" as warm to us means soft treble & wallowy bass & what an aged amp is. Transistors are usually used more in a design, some amps can really overuse them & then have to use lots of Feedback (NFB) to keep the volume correct in all the stages. A quality early Transistor amp from 1965-69 will still be designed on Valve ideals & will always sound "more valve-like" than one of 1972 onwards. Transistor sound is noted as an edgy glassy sound. This is not the transistors, but poor design, over-design & cost cutting. upgrade the cheap parts & you can tune out this edgy sound, but it might be impossible if the amp is overdesigned. A transistor power amp stage at it's best uses just 6-8 transistors, seeing some hideous overdesigned things with 15-20+ transistors is a joke & designed by a robot not a person thinking it sounds good. Some of these hideous amps sell for high prices to those equally unaware of the "real sound" of the early stuff. On the Sansui 3000A, an amp from as early as 1967 that we rate highly, several people have told us they bought one & are blown away by how wonderful it sounds, that lovely rich but strong detailed sound & a top phono stage too. It has the "valve-like" sound. The Valve is the more honest as it has less circuitry, the extra "edge" on even the best transistor amp is down to a lack of finesse in the design. One for upgradecers to deal with.
The difference A transistor amp will generally have 2-4 transistors for Tone-Flat and 6-10 for the Power Amp, this can sadly only blur the sound a little however you try to upgrade & focus it with sweeteners. A transistor amp could feasibly be built using the same amount of amplifying stages as a valve one though background noise without any NFB may reduce fidelity. The lowest transistor count amp we have is the Sansui 3000A from 1967 with 2 (per channel) in the Phono, 4 in the Tone-Flat amp, 6 in the Power amp including a transformer to make the PP stage. Valves done properly are free from user-fatigue with only the limits of the vintage vinyl we play music from the weakness. The Trio WX400-U has 2 for Phono, 1 for Tone & 4 for Power Amp, counting double triodes as two, this is the lowest count you'll ever find for a PP amp, though the Sansui 3000A uses the transformer instead of a transistor splitter. Those Single Ended Directly Heated 1920s 4w amps will count less but not be full audio range. The smoothness in the WX400U is as close as you'll get, but only once upgraded. Also we've found early UK made transistor amps like Ferrograph can sound rough too from average design with low spec parts, the UK hifi sound is not one we've liked therefore.
We are still waiting for someone to say A Valve amp sounds like a Transistor amp. So we'll do it as no-one else probably will: the Rogers HG88 MkIII we redesigned the power supply & recapped it & upgraded to perfection. But one problem couldn't be overcome without too much surgery. The Aux level inputs have a huge 220K resistor before the first valve which actually makes the Line level into the same volume as a MM Cartridge, ie 2V max through a 220K resistor to give about 20mV so they didn't need to add an extra valve for the phono. For all we done, the valve amp still sounded a bit grainy, just like a transistor amp can, for reducing the line level to Phono volume & then amplifying it up again. To bypass the 220k resistor & adjust the volume elsewhere did sort the roughness, but then it brought up a hum noise as impedances changed. An awful cost-cutting idea & it sounded... like a rough transistor amp. Nowhere else to go with our masterpiece beyond major surgery which isn't a good idea, so one lucky buyer has it now & must think they are in Audio Heaven.
Valves: They Get Hot. ↑
This may not be obvious to some, but an amp has glass valves with heaters inside as this is the principle of a Valve or Tube. In Hot Weather, not that England gets much of that these days, the heat can be a bit much. Sitting in front of the Trio WX-400U with all it's valves for Tuner & Power Amp, it gets a bit warm & the transformer paper & wax smells a little. The equivalent of the heat a lightbulb puts out multiplied by the number of output valves is the idea.
What A Good 100% On Spec Valve Amp Sounds Like ↑
Valves done right shows treble doesn't end in a gritty sharp edgy sound as many amps valve & transistor serve up, but it fully extends into a properly resolved sound in a way of comparing a rough worn knife blade to a sharp newly honed one. Bass can extend way low to have 20Hz audible though Valve Transformers can struggle with 10Hz & below. Midrange will be pure and clean. In fact exactly all you'd dream Hifi should ever be. Valves done right actually sounds much like Transistors done right, except Valves has sweeter harmonics & transistor has harsher ones which generally gives Transistors an edge to the sound that may make you think the Valve one isn't as detailed in the treble, as that slight edgy treble sound is always there. But Valve amps need maintenance, revalving & then rebiasing & to number each valve is a must as you take them out to dust or service more.
What A Tired Off-Spec Well Used Valve Amp Sounds Like ↑
This is the Sound most will think is "Valve Sound". Off spec aged "Old Knackered Amp" valve sound says we. It can sound appealing in a Retro way as our 1932 Pye G/RG radiogram can, but it's not Hifi & only by about 1949 did a sense of modern Hifi begin as NFB was introduced. A slow wallowy overloud bass, the often noted "Warm Sound" to us means a rich sound of not too much detail & treble well rolled off, just like most old Valve radios will sound. "Warm Sound" actually does not have any crisp detail. The Bass will be bandwidth limited as per the old low value capacitors & the power supply will usually be very low spec so it couldn't cope with fast treble detail anyway & any Bass would cause Voltage Wobble as we found on our Trio WX-400U when original. The heavy bass limiting actually causes roughness in the treble, so it'll be rolled off to hide this. To upgrade an old Valve amp is worthwhile within limits but bear in mind the power supply is hopeless as it is with such small values & then the problem of multi terminal capacitors, ie 3x 16µf in one can. 16µf is a hopeless value in a valve amp, the Rogers only use 100µf. Look inside your transistor amp & see what values it uses. Find an old 1940s radio with a big speaker in & play a station with male speaking voices. Hear how boomy & coloured it sounds. It has Retro appeal which is why buyers like the prettier ones. Old Jukeboxes sound like this, but those under 30 often like it loud & in your face, only later does crisp treble quality & finesse matter. This is because the design is purposely limited to keep it cheap & be "good enough" for the customer of the day. If you were to try to upgrade a 1940s radio to sound modern you'll quickly find how basic the design is & you'll only reveal how poor it is even by adding a tweeter. Transitional buyers in the late 1950s complained of this.
We've restored 4x 1960s Valve amps ↑
the best sounding in terms of musical value is the Trio WX-400U, the Sansui 500A had issues, the Rogers HG88 Mk III could be made better without the awful high input resistors & then the runt of the litter was the Armstrong 221 which was pretty average, pity those buying the receiver versions. We've also had Rogers Cadet III in the early 2 part & later one parts as well as valve radiograms. Our 1932 Pye GR/G radiogram is working now if 81 year old design now, it got another Pye G radio bought & innards swapped over, the radio version had all original parts & works fine. The gram one had a replacement transformer & visible capacitors & resistors changed, which was less pleasing. We rebuilt our own valve amps to our own design after gutting the insides, that was in 2008 & with the odd upgrade since they are still used daily. We know valve amps & took our ideas for design from various valve amp circuits. Valve amps are the best way to get the best sound. No question of it. But it's not easy & it's not cheap.
Budget Types Of Output Valve ↑
These are Early ICs of sorts. The Rogers range uses ECL86 as does the Armstrong 200 range. These are a compromise TV style valve with a preamp valve, as in ECC, and the output valve, the EL so why it's an ECL as a combination. This doesn't seem such a good idea, a bit like the IC of it's time having two very different stages in one thing. The Trio WX-400U uses EL84 output valves of the same size but uses ECC83 for the splitter stage, so uses two valves where the Rogers & Armstrong use one to save space. You can still buy these valves easily as many amps use them & old Guitar Valve Amps will use them too so there is a demand. But whether the ECL combo ones are of less quality to an ECC and EL pair, look at a circuit: the HG88 MkIII runs each half at different voltages for a start. More about types of valve you can read elsewhere. For best quality therefore, look beyond the ECLs to the amps that are beyond the midprice status of their day, Rogers were not sold as expensive items compared to others in 1966, for the premium items Lowther, Quad & Radford were a preamp & monoblocks or one main amp, not a small box with the lot in. EF86 found in the Quad IIs was from 1954 & is a single stage preamp valve. The ECL86 triode-pentode was only released in 1961 to cater for the budget amplifiers & portables. Others like the two stage preamp ECC81-83 range were 1951-54, EL34 and EL 84 in 1953-54 too as the EZ81 rectifier. Bigger valves like KT66 were 1947, KT88 as late as 1956 and the huge 20w PX25 the Decolas use is from 1925-32. More info on the Valve Museum site and others.
"But old valve amps 'as-is' are mostly not very sophisticated for modern use beyond being aged & need much very advanced reworking to sound as good as you'd want them to. It can be done to a certain degree or to perfection if you know design. To keep any Vintage Amp still looking as Vintage as possible is important to keep it's value & increase it."
Old Valve amps have Fixed Bias. This means that the Waveform halves in a Push-Pull amp with older valves will have a slightly inequal output bringing distortion which can give an edgy sound if opened up fully on the treble, which is why they rolled the treble off, as there is no way to adjust or properly Bias them. Surgery can add these parts in, but advanced work to do. But there is an easier Solution: If yours sound too edgy, go buy some new matched valves as this will help this problem a lot. Valve sellers sell matched quads or 4 valves at no extra cost simply as Guitar amp users buy these too. New Valve Amps of sufficient quality are very expensive: Valves done right will never be cheap. Those Chinese imports may give a taste of Valve Hifi, but we'd not bother with one as we do not rate them worthwhile. If you don't fancy the dark world of Valves, 1970s Hifi is your best option & we've written loads on that. This page is pro valve for the great sound & anti valve as for the buyer it is a very hard scene to get into unless a sharp buy or an aim of spending £2000 is in mind. Valve amps need regular maintenance & new output valves bought every 2 years if you use the amp 8 hours a day like we used to.
You Want Us To Review Top Quality Modern Valves? ↑
We'd love to, but look how many Transistor amps we've covered so far. To go through all the Modern "High End" (ie expensive) modern valve designs would be enjoyable, but there is a problem or two: we know the price of our Tube Technology when they were still available & the preamps too is heading into New Car money. The odds of finding shops desperate to unload ex-demo ones that they can't shift to flat-earthers is probably one & we bought those in about 1998. To buy Valve Amps online on Hifi sites paying top ££ or ebay & talking severe risks is the only way. For the fact we'd always want to improve them as we often do with our favoured Transistor amps means you couldn't buy to try, or would need a Millionaire bankrolling you who didn't mind you'd have the cheek to "fiddle" with expensive Hifi. No, only items that are ours we can do that with & to just try any worthy amp isn't a 3 day affair, it's a courtship, some of our best amps have been here months & only go once we go as far as we can or want to slim the pack down. So really until the economy wakes up & money is flowing free like 2000, a long time ago now isn't it, we have to wait to try bolder equipment like the McIntosh we were trying in 2003. Then we'll have a go at importing McIntosh & Fisher early ones that are nagging us to try them, but not yet.
Valves Do Need Replacing ↑
If you've got a Valve amp, keep a few spares as they do fail in several ways. They get noisy, hum can be unacceptably high with one valve on one amp but be OK on another amp depending how hot it's run in it's design. Obvious lack of the inside silvering as a whitening or no traces means the valve has chipped on the top or the pins being bent cracked the base & is now with no vacuum & is useless. Bin it & buy a new one though rare ones can be repaired. The emission can go low meaning poor sound or low volume. If the amp sounds distorted where it shouldn't be swap a valve & it may correct the problem. If you can't find anything at all wrong with the circuit, consider swapping valves around, even L to R to spot a duff one. Valves get hot like a lightbulb so protect hands. Be sure to switch the amp off before swapping too. Guitar users after a heavy session can find the valves failing the next day though a bit of DIY can get them going again. We've repaired our EL34s many times, not a case of buying cheap ones but these have been used a lot longer in time if not on as much as when we had to replace more often. If the silvering fades it may suggest the valve is weak. Other problems like the body of the valve glowing deep red, not just the orange heater show problems that can trash a transformer. Preamp valves you can buy good tested old used Mullards & get years of use, but Power Amp valves best buy new ones. We use old used Mullards in our preamp stages & the input stages on the power amp. Don't believe the swapping a high gain ECC83 for an ECC81 or ECC82 to get more gain as the design will be different & not be correct. USA valves like 12AX7 are the same as ECC83 though some modern altered versions with numbers nearly the same like E83CC. 7025 is the industrial version & there are Military Spec too. Which you want depends on what you've read. Then there are the Tube Rollers who claim to hear things sound different, based on tiny design variations. We've never bothered much beyond seeing it's not worth bothering.
Bad Ideas in Valve Rolling can Kill an Amp ↑
If you must roll valves, make sure your Amp is not Fixed Bias, but has adjust pots which will need setting right to stop destruction. If the Bias is fixed, only use the same valves, no 'improved' ones unless you can read the current & adjust the resistors. To swap between ECC81 aka 12AT7, ECC82 aka 12AU7 & ECC83 aka 12AX7 is still not a perfect thing to do, as these all have different characteristics as the spec sheets show, it's not just the same valve with different gain, the Anode resistor values vary as does the Cathode one. To go beyond these is even more a dangerous game. Sylvania GE5814A supposedly a Military version of the ECC82 or Mullard M8137-CV4004 are supposedly ECC83s, this is said only by the sellers & knowing the JAN valves are rubbish, we suspect a big stock of useless valves is being passed off to the unwary. Don't confuse JAN name with new JJ valves, those are good ones we use. But without digging out the spec sheets, the current draw can be very different. If a circuit is designed to perfection with ECC82 or ECC83 why mess around with ones 'that are better equivalents' if only said to be by sellers? See the Trio WX400U section for why valve rolling is not a good idea.
Valve Testing & Reusing ↑
The above shows a Sovtek ECC83 (7025 12AX7WA) dated 12 07 like ones you buy new still, can go weak quite easily. There does seem to be a strong case of using a valve from new in one place only. We have a few assorted valves from recent & ago but it seems that a valve appears to age to suit it's circuit & characteristics & used elsewhere it might be far from it's new spec. This makes us wonder as we have some Yellow Mullard ECC81s that we used regularly but found them to be equally weak when tried in the Trio compared to more recent Edicron ones. In our experience the Edicrons were reliable, Svetlana on EL34s were unreliable on the heater pins sometimes, the Sovtek one went weak quite quickly though it could have got damaged in the ill-fated Sansui 500A that ruined too many EL84Ms so we'll ignore that as Sovtek are reliable & JJ ones as EL84s in the Trio are fine. The way to test valves is to use a valve meter, years ago we bought the old Avo curved top ones with the roll wheels on top but on looking inside they are so old, how can you be sure they are calibrated right so we sold both on. The other way is to have a few valves & swap amid them, reading off the gain using a test signal. We done this with some ECC83s and in our test we got as high as 2.3v and as low as 1.1v showing emissions strong to weak & the gain will be affected as will Tone & Phono EQ. So is it wise to buy preused Mullard, Tungsram or other vintage valves? Old circuits ran the circuits 'cool' so the valves will have had an easy life, but still have 'grown into their original circuits'. On first getting Mullards over ten years ago they were preferred to the Edicrons as they sounded smoother. But on trying the ECC81s we've had that long, they were soft sounding with attack weaker than a newer valve. Will be be buying more vintage Mullards as used on ebay for £20-40 each? No. Will be be spending £400 to buy a new valve tester? Hardly worth it when new valves of ones we use not expensive to buy. To that opinion, what is the appeal of buying old used valves with no idea of where they were used or how much used? When we bought ours years ago the appeal of the Mullard original but we've since found Mullard made rubbishy transistors that aged badly like the germanium AD140 & the BC147-149 range that sound very rough now. Mullard certainly made quality valves but these are now nearing 65 years old, the Yellow ones are mid 1960s & the white styles are later into the 1970s. They are not a wise buy is our opinion now. To buy Mullard EL34s is foolish we though back then, beyond the high cost, as the new ones if used all day only lasted 2 years before getting weak so bias couldn't be set right.
Good Valve Amps ↑
Ones made post 1980 with Guitar Amp techniques & higher power & more sophisticated design, ability to adjust bias & better power supplies. This is what VALVE AMPS are about. 1920s Audio ideas brought up to date & done properly. No mass market companies involved & those cheap China imports aren't much good. You need to spend big money sadly. Valve amps done properly aren't cheap. We use the early Tube Technology yet the design inside is all ours, only the case & transformers are theirs. Their design is OK but ordinary to the point they sounded boring as such a safe design.
Classic Valve Amp: Heathkit S-88 and S-99 ↑
The S-88 was introduced in the 1959 HFYB and is an 8w Stereo integrated amplifier. Distortion 0.1%. Price £25.5.6 with a 52 page book & wall posters to make it easy for anyone, if it involves 300v. The S-99 was new in the 1963 HFYB and is 9w, £27.19.6 if only for 3 ohm or 15 ohm so 8 ohm must use the 15 ohm tap, not connect - to the 3 ohm & the + to the 15ohm as one person though was a good idea. The S88 has the inside different to the S99 which has the preamp horizontally. Pretty ahead of it's time as very few integrateds were around & compared to others like the Pye Mozart this is quite a decent amp. But that's the trouble today, they are now very old & the fact it was a kit means you rely on some unknown person's work & odds are it'll not work as the capacitors will be bad by now, especially the plastic axial ones. Both S-88 & S-99 look the same & hard to find with the perspex uncracked or not flaking. But that's only the start of the problems, the design is very cramped making upgrading it tough. The one we had years ago hummed in the audio & physically with a buzzy transformer, possibly not constructed right or just aged. How to rebuild this would involve getting the original book & just about doing it all again though space is too limited to upgrade it much. But it's not a cheap kit thing, good sized transformers, ECC83 & ECL86 valves (on the S-99) are easy enough to replace but it does have a valve rectifier limiting the capacitor used after though you could rebuild it as you choose & use diodes, if the HT will be higher then. But is it really worth rebuilding? If you are looking for the best Hifi sound, forget it, but if you want a starter valve amp & don't fancy paying the excessive prices people ask for Rogers Cadet III, try a Heathkit, recap it just to the original spec on those small plastic capacitors & give it a try. Don't use it for too long if you want to give it a quick try first as it can fail. Seen these go for under £50 so you can not be too bothered if it doesn't work out. In July 2014 a rough rusty but all original one made a hefty £151. For the fact it needs a total rebuild & respray the buyer must be a little naive as these aren't very high spec & never will be even with how we'd upgrade. Not worth the effort says we.
Valve Brands. ↑
A quick opinion on various valves we've had. New valves be they newly made or NOS are the ones to get. Buying used Mullards may be appealing, but the valve will have burnt into the original circuit & your different spec may bring erratic results, although you can probably 're-work' them into your circuit, it's like buying used clothing, it'll never be as good as new. Therefore used Yellow or White early Mullards can sound smoother than new ones, but it might be as they are aged despite reading good. To buy New Valves, you should expect ones with Test Results & be very closely matched within 3%. These read Gain (A), Transconductance (gm) and Noise level (dBv), read elsewhere for more. Some brands are reliable & others you get the odd duff one, but that suggests more the seller doesn't test them & run them in first. On buying Svetlana Winged C EL34s, the last lot we needed 8 so bought 10, 1 went bad quickly glowing red for no reason & the other spare never got used. These are generally OK but the pin soldering on one was poor so the Heaters stopped working until the valve got repaired, which is a worry as other pins might go bad, but the Heater is a heavier current. There are several brands that have Gold in their name, to hint they are better. We've not found these so good, Golden Dragon, Golden Aero & Gold Lion. The ones we liked least were the JAN ones that one seller has a huge amount of but no-one else does. The ones Maplin used to sell, Edicron & Ei were reliable, if Edicron looked a better product. The JJ valves have been reliable too & they sound much cleaner tha old Mullards. Vintage Mullard & Matsushita ones being good too, being found in their original amps. A problem with many valves is the print wears off, in use or kept loose in a box so you have a valve that might be an ECC81, 82 or 83, or maybe not. Sadly to just throw them out is the result rather than just guess. Old valves that came with Quad & Rogers amps were usually still good. Those with Trio & Sansui valve amps were not so good & needed replacing, though the amount of use & amp condition will affect things. Buying new EL34s today from one ebay seller, the choice is the Svetlana, JJ, Electro-Harmonics or Tungsol. You can read on Guitar forums, far more aware than Hifi forums, that each has it's own Tone & Overdrive, but guitar users want distortion. For Hifi use, generally there is little in it beyond trusting the brand. We used Sveltana before & find the JJ ECCs good so will try JJ EL34s when we need to revalve. Revalving is only really needed if the valve fails, beyond the pin repair or the Bias setting is wrong & can't be put in spec as too low or too high readings. One of the Svetlana EL34s has barely any silvered getter on the glass but it reads fine so it's still good. The getter is to absorb moisture in the valve & the one here might have had manufacturing issues but survives. Only when the getter silvered mirror part goes white the valve is dead or if the insides get too browned looking revealing the valve is worn out & emission will be weak. This may not reveal itself until you play it louder. We bought one EL34 set from one seller & several locking pins broke off, not just one from being clumsy, but several, like they were not straight & broke off being put in & cracked when hot. Rubbish. If they are that bad to not even get the spec right, don't risk using them. Most new valves are made in Russia or the old Czechoslovakia for some reason. There are various meters to test valves, we got the AVO old ones but being so old they might be off spec anyway. Beyond paying £500+ for a modern valve tester, at the price new ones are, easier to check Biasing & use your ears to decide. Just buy new ones that have been run in & tested, preferrably from a Guitar Amp seller who will be selling lots of valves to guitar guys who will know the brands better than Hifi users as they'll get through them faster.
But It's Only Ten Watts. ↑
Looking at the HFYB specs on valve amps & few are over 15w RMS, even Rogers HG88 early ones are only 8w, Cadet III is 10w. They're not going to sound like you want, using them with the old components even if it does basically work, which doesn't mean it'll work next time though. They'll be sweet, soft, old fashioned but ultimately you'll have heard better. But you're not hearing them at their best, valves done properly are as detailed as the best Transistor amps but with the superior Valve sound & effortless liquid Treble you'll be in love with the rest of your Days. We didn't feel we've done these more modest Valve amps properly so got the Trio WX400U that took over two years to get right as we learned more, a Armstrong 221 in a custom case cheap that we've now fully recapped but sold on fast as it was mediocre & a Rogers Cadet III that we couldn't resist for a mere £30 on ebay, but then developed problems on the connecting cable plug so parted it out. Valve watts mean higher clean sine voltage than equivalent Transistor watts as our Watts Are a Con page shows. 10w on the Armstrong through our headphone box plays as loud as a 35w one, simple circuits mean less is lost. Only deep loud bass can clip them out being the only limitation, though even at 10w the level is still more than you'd expect if only used to Transistors.
The first Valve Amp we ever had ↑
we just saw on ebay: a 3w Mullard Stern's Mono amp, basic 3 valves EL84 output, EZ81 rectifier & EF86 output, a Home Constructor type amp. "Stern's Mullard 3 valve 3 watt Amplifier". 3w of pure Class A like some early Portable record players & the sound is very pleasing in a small cute way. Can't see it on this one, but ours we had 20 years ago had a TV aerial coax input. The sound was so pleasing even at 3w compared to amps we had that it had us looking out for others that came into the same Charity shop. In those days it was worthwhile knowing & helping the staff out to get the best stuff. It's where we got our first Underwood Typewriter for a fiver too, as well as that big WW2 era Tannoy speaker which had a sound so unlike the big Fane speakers we used at the time. Then a Rogers Cadet III arrived & we, er, had fun with that. The first ever valve record player we used often was a long Pye Stereo Black Box coffee table type record player, the one with the piano lacquer wood & the Garrard player. We actually got one from that same Charity shop though the transformer burned up melting the tar which sadly meant the end of it.
Is It Worth Rebuilding a Rogers Valve Amp? ↑
Rogers in the 1959 HFYB have three ranges, the 20w Senior, the 14w Junior & the 6w Cadet and then the 8w HG88 Mk I. So the Cadet is the lowest spec one yet as Rogers (and Leak) realised there is more money selling cheaper gear mass produced than the better quality Senior, it's why the Cadet & the later Ravensbourne-Ravensbrook are really only budget priced amps, the valve ones still being decent but the transistor ones are pretty lousy. After having had good success with the Trio, as well as doing well on the Rogers HG 88 III until realising there was no tidy way around the high input resistor for Aux, to look further is an idea. The obvious one is the Rogers Cadet III as well as the higher spec Rogers HG88 Mk III. These sell often on ebay though we have noticed it's the same one again sometimes. We'll think aloud on this one as we look at photos of one, so you get the idea on it. We've had three of these Cadets before, the last was about 7 years ago for a one case one & a year ago for the two part one. Had three of the HG88 including a Mk II. The power supply uses four capacitors which are in three cans on the earlier ones as was typical, though later ones have four cans. To upgrade to our spec would involve needing four cans so actually isn't so difficult even if it's the three can one case version. The very early pre & power combo that fits in the case only has three capacitors & the joining cable is highly risky as it ages so sparks. The next thing is the output transformers. These are very tiny compared to the size of ones used in Receivers quoted as 10w as the Cadet III gets rated. The size of transformer & why it needs to be a bigger size is how it deals with the low bass, not that the original design lets much bass through. The HG88III we redone was 15w & had much bigger transformers & coped well with our upgrade deep bass, how the tiny Cadet ones would cope could be a problem. As with the HG88III the ventilation is poor so they get very hot. Rebuilding means lots of soldering & newbies will overheat joints in desoldering & find the black plastic melts leaving the wiring posts in trouble. The thing is you may know how to do it properly, but these Rogers amps have been used & messed with over the decades when other amps got a few years use & put away. To try to redo bad work is often very hard. The Cadet has Mullard mustard capacitors you can still buy as NOS but the old values limit the sound hugely as we found with the HG88III. To work on the one case version is not hard as it's nicely laid out as is the HG88. Earlier Cadet & HG88 may not upgrade as well as the III versions & the earlier ones have Valve rectifiers that you can read more on elsewhere. We know one shop charges £300 to service a Cadet III, he has shop overheads. To fully rebuild a Cadet involves quite a bit of redesign to our no-limits ideals, but whether the tiny transformers could cope with deep bass is the gamble. No they couldn't is our opinion & to limit the design is a bit pointless. The Cadet III can't take a full upgrade, but if rebuilt to a more limited ideal, it'll give a pleasing sound to begin you on your Valve journey. The Cadet/HG88 III amps have a few cheapout features that limit fidelity. Instead of using a separate Phono & Aux-Tape valve, it uses only the Phono one & a large 220K resistor is what your CD sound goes through. It makes it sound grainy & is only a compromise. Also the ECL86 output valves are a compromise like an IC as they contain a preamp stage & power output stage in one triode-pentode valve. Then the ECC807 valves are obsolete, you could change them to an ECC81-82-83 but pins differ as does the design. The balance control isn't like a modern amp as Sony introuduced in 1965, but instead adds gain with a turn to left/right. Together with tightly spaced Phono inputs that do take modern cables, the other problem is the Speaker connectors. They are just a line of screws. Do not alter these. We've seen plenty of bad replacements, but all you need to do is use a fork connector on a cable, or buy those gold plated McIntosh-Marantz advertised L shape blocks that allow a 4mm plug or bare cable to be attached. These amps don't have Headphone sockets either but you can make a headphone box & connect it to the speaker outputs, it's how we used ours. Never fit one into the case drilling holes as you'll sell it one day & buyers don't like that. There are enough of these amps around to buy a spares amp or find knobs & cases for these, though look for the ones with bright gold coloured (cadmium plated) metal inside as these are the ones that saw little use. Any ones a bit darkened or rusty inside should still be ok if less attractive. Prices vary hugely & despite us knowing these well, we know they really do need a full rebuild as those dried out power supply capacitors will fail sooner or later. We've cut ones open & they are bone dry. A capacitor for Hifi (not the modern solid ones used for computers) requires wetness to work & for the fact these see 400v a failure may knock out a transformer. We've done our rebuilding of an HG88III now so won't be trying any more Rogers. The Rogers tuner that matches the III range is a very basic type only suited to receiving the few BBC stations at the time as it has no typical Tuning Dial, pretty useless sadly. The III versions have silicon diode rectifiers, the earlier ones before the 1963 ranges have valve-tube rectifiers. A nice vintage touch, but they are known to sound 'slow' and the problem of putting an unlimited power supply isn't possible. It seems only the 1963-1967 valve amplifier or receiver ranges are the best. Also the early Phono stage may be for a Ceramic-Crystal type cartridge, not all had the MM type Phono input though the Rogers uses adaptors & the Trio below has both MM & Ceramic. Armstrong, in their usual cheapo way used transistors on their 1964 range or just a Ceramic Phono input. If you want us to fully rebuild your Rogers amp, see our Upgrades link at the page top.
The Story of the 1963 Trio WX-400U ↑
This was the first vintage Valve amp we got since the Rogers Cadet III in the early 1990s. One of the very first Stereo FM Receivers from Nov 1963 together with Pioneer & a few months later Sansui. One of the best looking receivers ever, before the dull grey metal cases. We got ours in mid 2011 at a nice price. This amp was still listed in the 1967/68 Hifi Yearbook & will have looked very old fashioned, but now retro means it looks way cooler than any. Rated 10w which based on the Rogers Cadet III is about 35w in transistor amp terms. Fixed bias on the WX400 but to buy a matched quad will solve that, unless you fancy adding in pots to adjust. There appear to be minor changes as the run progresses with the HFYB one having the glass different on the earliest ones & a DIN socket on the later ones. The original manual we see states 20w music power, 18w continuous rating which makes more sense than 10w, though 10w is about the lot with the original spec. The Sansui 1000 from 1963 or 1000A from 1967 are touted as the best valve receiver made, but we've found the 1000s have a remarkably high 150k resistance (3x 50k) between Tone & Power Amp for the High Filter to ground, the Sansui 500A has a better High Filter in a NFB loop & WX400 wisely has no high filter, the Noise Filter is only for FM. The trouble straight away was there was no available circuit diagram until a kind reader with the same amp sent us it, so we'll share it. Download the pdf here. Opens in a new tab/page. To do these hardwired amps is way more advanced than a nice circuit board with numbered components, here you need to figure out which ones are which & be able to recap & upgrade it properly. Photos & numbering things helps. We had many problems with the power supply being so hopeless if the caps were still correct values. The grey 0.05µf 600v capacitors age badly & we had failed power resistors too & until geting the circuit it's not good to guess. After recapping the volume was pretty feeble though it was way improved. A knowledge of typical valve design shows where to improve. The WX-400 has ECC83 & EL84 valves which are still made & much better than the Sansui 500A with it's obsolete 7189A valves though EL84M should be ok. But the FM Stereo never worked which took many tries to find a fault & then to adjust it to bring it to sound wonderful. The original amp we did try & it sounded lousy, wobbly voltages, bass clipped out a very tired sounding amp. It was a shadow of what it is now & at last we're not selling the thing after many give-ups. But the reward for presistence is an amp that now sounds better than any other valve amp which still sounded a little grainy & probably needs £250 of new valves to buy from USA with extra shipping & import charges. The trouble with the WX400 is the fact it's a 1963 design gets you thinking it's too old to be good enough. There are a few cheats in the design to cover up for poor design, out they went. One physical part that is poor on the WX400 is the balance control which will be replaced as it's substandard & affects the fidelity, though bypass wiring it over will do. The modern type of dual-pot balance with only half of the turn affecting must only have been invented with 1965 Sony TA-1120 as Rogers HG88 Mk III still used the ground in middle type & the Cadet III the strange louder if turned fully L or R type. The WX400 tone stage isn't very good either so we found a better design in an Ampeg guitar amp & used the values giving a far better sound using the original pot values. The 5 pairs of Phono sockets on the back are useless as found on the Sansui 500A & some early Sansui 3000A. Then the headphone resistor is way too low a value, to use a typical value again brings it more modern. It needs a hell of a lot of modernising really, way more than any sane person should consider attempting, but the reward is certainly worth it & the best looking receiver looks are another reward. There is one valve after the aux, then straight onto the power amp one & then the PP output pair, perhaps the lowest count of amplification possible. But since having got this amp, we done the Armstrong one & thought it was lousy, then got the Rogers Cadet III two parter that sounded so nice as barely used but it had issues after some use. Then we got the Rogers HG88 Mk III & done absolutely everything to bring it to sounding it's best, but ultimately limited by input resistor lazy design. So the WX400 got another try. Our valve life started with a Rogers Cadet III we done too much with in the days they were of no value & several years later rebuilt our Tube Technology amps from sounding nice but boring into the fidelity that we crave. Tackling an FM Valve tuner was done first fully with the WX400 tuner which was actually nearly done from before & works fine on germanium diodes still, so they stay. The Phono stage got several hours use playing a batch of 45s, just needs a few minor changes. But it does show the slight mains noise that can't be fully tuned out by the Hum adjuster so it requires some power supply improvement, seeing work you done quite a while ago & seeing it could be bettered is the learning curve. Plenty more to learn if you were building valve amps from scratch, selecting power & output transformers & positioning things to cut hum. Also Phono lacks a true Mono but the Blend control can be altered. The noise & hum balance is because all Heaters except the 2 phono are AC. Hum Balance is for balancing the Tone & the Output Cathodes which is a bit poor for today, but as it's there we'll do better on other things, ripping sections out because you don't like the idea is going too far on Vintage. The amp sounds a little weak until it settles after about 5 minutes & then it sounds wonderful, but the less great sound from on is another one to find out why. Upgrading the power supply further, which is not easy with the space & positioning limits, the background mainsy noise is reduced to about a third but the sound is now even cleaner & tighter. The sound is the cleanest we've ever heard in any of our amp compares & how it compares to or shows up our main valve amp will be interesting. Was it worth the effort in the end? Over two years later to get to a level we are pleased with. Yes, but the getting there was remarkably tough & we gave up several times & tried to flog it. But we know very few understand valves to see what it was already was even the last time before getting FM Stereo working. There is a risky therapy in offering things for sale as it gets you on edge a bit whether really to sell it or try it again. It got looked again a week before we fixed then the next weekend it got done. Trouble is someone might have bought it & one nearly did from the last time we listed it but bought the TK-140 instead seeing it was 'gone'. For fitting a few parts to the Sansui 3000A it got tried with the same tracks the next day. It was very clean but the splashy treble of transistors & their requirements was so different after the WX400. Splashy treble from how the design is making the treble ring on a cymbals hit that the valves played as sweet as you want. If the 3000A was without it's limiters it'd pick up RF & overheat. The WX400 plays it beautifully with treble pin-sharp in it's accuracy with no artifact or limitation beyond the music track fidelity, all our tracks are recorded with a valve phono naturally. Mono tracks sound their best too, on transistors we've not liked Mono so much as it was too mid-head on headphones, with valves it's just right. The game of finding the Best Hifi is over. It's not in Transistors, if about 75% of the valve's beauty is attainable with transistors, the nth degree isn't (so far). Not that this is going on the Top Amps page, but in a compare Yamaha CR-1000 sounded the best followed by Sansui 3000A but Sony STR-6120 sounded way too harsh on upper mid-treble when the sweetness of valves is in the mind. But all transistor ones sounded a bit cardboard compared to the 'easy' valve sound as the bass is restricted. Looking at the Sansui 500A again we see the Trio has just a bit more space for bigger PS caps. The Trio has low NFB with 27K compared 12K in the original Sansui design, the HG88III & our main amp uses 4.7K, Cadet III 6.8k, a later Trio WX-1100U uses 5.6K. Explains why it sounds so fresh. Lower resistor means more NFB, but there are other issues like impedance that stop one resistor giving the same db of NFB on any amp. The Tone, 2x Phono and 2x Driver-Splitter all need DC heaters to lose the mainsy buzz and the hiss & hum to a very low level matching a transistor amp. It actually is possible to get a useable ±DC voltage for Heater & Bias and then fine tune the Bias to the optimal ratings for the Anode voltage specs. The original design used Cathode bias before the negative voltage bias was used. It actually now seems like it was designed to be better than the retailed version by having the extra tap & dumbed down as ahead of it's time, hearing the sound that comes out of a 1963 amp once rebuilt is insane. Heaters are best done as DC which is why so few old amps do DC heaters except how this attempted it. Well worth doing DC heaters & bias right to lose the noise, though if you'll only play it loud, you might not be bothered. It now sounds explosive in it's dynamics from a whisper to a scream is very unusual to hear from Hifi, but a far cry from exploding capacitors & fried resistors when we got it. Next comes tuning the Phono stage, as usual Phono is never as crisp as FM but can be altered & it sounds very good. The last problem is the Mono-Stereo blend control. When new it would have gone to full Mono right, but like the Volume control is with being noisy, inspection reveals the Blend pot is worn out with 20k ohm+ instead of zero. To fit a new pot may seem easy but the spline type is unfindable & with such a long rod on it. We had an Alps mode selector from an Eagle transistor amp that fits perfectly & easier to use, you'd only use Blend for Mono, Stereo & half blend & the switch has 3 positions. Did try a pot using part of the original one but it isn't a good design as only actually blends on a tiny amount of the rotation as well as other issues. To then sort out the lousy Balance control & the noisy Volume pot which may be worn too as the noise is the same place & the blend one was quite weakly made. Do Alps Blue do a 1M loudness pot with a spline rod? Does this amp sound a challenge at all? There is still more to do in the Trio, fine tuning the valves to ideals pays off hugely with a sound that sounds unreal.
Trio WX400U on Tannoy 15" Golds. ↑
This was the next thing to try, we've never played any of our transistor amps since starting these pages on these speakers & actually no other amp has used them in over a decade. How did we deal with the screw connectors on the Speaker outputs? Easy. You can by McIntosh-Marantz 4mm banana & binding post L shaped blocks on ebay & we had some from long ago, probably for using the Heathkit S88 or S99 years ago trying it for a computer amp though it was very lousy & hummed too much. Speakers plugged in using cables we had also from long ago, check the Impedance switch 'Imp Selector' is to 8 ohms & the Phasing Switch on ours is correct pointing to the 16 ohm side, to set when you first hear music. It sounded excellent. Actually it played louder through the Speakers as per the Volume Control position so we can change the Headphone Socket resistor to match. Only very minor background noise can be heard on speakers though the Volume control is a bit noisy still. A minor adjust once the Headphone socket is altered will lose that tiny bit of noise as it'll be matched better, with a different resistor value we now hear the noise as heard with the speaker. The solidness of voices on TV was pleasing, treble was fully extended without any treble edge. Bass similarly was only there when the program has it. Very high quality sound. It showed our own Valve system up as being now in need of a redo, we did build it 6 years ago which is a very long time in the progress & amp rebuilding we've done since. But the challenge is now to match the Trio on the 100w system. As is usual with this amp, it sits for a while & then gets new ideas tried on it & it wins again. Others come along & better it but then inspiration comes along & the Trio is back sounding great. But as of typing this after getting the Tape Head Sony STR-6120 back again & having it sound great, the Trio just adds further transparency to the sound as valves do with less circuitry. Various changes give it extra kick to sound much louder than before if no need to add an extra preamp valve as was being a maybe. But then further compares & even using Aux into the Xtal Phono input, it has a 470K resistor, a small capacitor & needs Bass set to min so sounds rough but suggests more gain is required to match the gain of the Tuner & Phono which can get the rest tuned a bit finer instead of being at the limits. Ideally to use the last Tuner stage valve but it'd need extra switch wafers & a lot of wiring or just add an extra valve in somewhere but having to cut a hole in the chassis.
Living with the Trio WX400U. The amps gets used quite a lot but to use it on the main speakers happens less. To use it as your main amp is probably not ideal for the amount of heat. 21 valves chuck out a lot of heat if only 5 are the main audio ones, the Tuner valves are always on & ready. The top lid gets very warm if still touchable but not for too long. No putting LPs on top as will have happened with these valve receivers. The transformer being from 1963 does smell a bit & it is a little noticeable in the room. It could be sealed somehow though disconnecting it all to take out would be probably not a good idea.
The Unexpected Return of the Trio WX400U: A Cautionary Tale. Well, we never thought we'd see this again, but after being sold 6 months later it's back. Sort of wish it wasn't as it's a bit sad now. The courier in the Drop Off shop did just that very hard & the case bounced so the rivets that hold on the centre chrome strip hit the glass so it needs a new glass made. The reason it got returned is down to careless Tube Rolling. The WX-400U is fixed bias so uses fixed resistors to accurately bias the 4x EL84 to a suitable bias. This we'd done & checked many times, it was steady & reliable. The driver & splitter is marked as an ECC83 though we found an ECC82 better suited. Both drew the right amount of current in our circuit. But sadly it got tube-rolled using those "equivalent" types some sellers have a stash of & make out they are as good "or better" than the ECC81-82-83 family of valves. But our design was for either ECC82 or ECC83, so put in a valve that draws too much or too little current based on the circuit & the EL84 bias goes way high. The JJ EL84s we'd used in the amp for 3 years with no problems ever now had the red paint burnt to faded orange. Putting some new ones in just to see what happened as the voltages were there & the output transformers still read right, the new EL84s started glowing red on the outer grey anode part. Bias was almost double what it was & all the resistors were good if what saved it was we put a cathode resistor to read bias off, the amp never had one before. The resistor that blew was 5 watts and calculations of current to do that are over 20A. The EL84 valve bases were discoloured from the heat & some wiring plastic had shrunk back. The mains transformer smelt of burnt caramel-coffee as you get with speaker coils that burn out, the failing resistor at least saved them as voltages are still right. But the fact is the overheating has altered the transformer characteristics. a transformer is layers of laminations with enamelled copper wire wrapped round to a precise design, but as when you anneal metal, it can be made soft or hard as it changes the structure of it. How hot the valves & transformer got is unknown, if the metal lid & yellow wire from the antenna aren't scorched, there is a chance it failed within an hour. The trouble is, what are the transformers like after such overheating? We've had an amplifier that got serious abuse by the user leaving the transformer so hot it oozed the potting compound resin out & melted ABS plastic, this will have been at least 200°C but after cleaning it up, beyond a bit of a smell, it was still good. Here the mains transformer is open at the base showing the windings & paper but no wax in it to ooze. The facts are it seriously overheated also to knock out a 5w resistor & cook the valves severely, it could have reached 300°C even? The result is now bias is now double on all new valves because the transformers have altered. Would it ever be reliable again? Here it runs on 400v for it's 10-12w & to mess around with high voltages is a lot more risky than the other overheated transformer running at 50v that had the potting to save it. The copper windings are enamel coated. Looking at modern transformer wire it can take 180-200°C levels of overheating, the receded wires on the valve bases stay put at soldering iron temperature but you can stretch them back so if kept hot as the iron for too long they will recede. Only by prolonged testing after altering the circuit would it be known if it was still safe or a day away from failing? You can imagine what shorting or breaking transformers can do, we'll not scare, but sadly it'd take a lot to trust it again on speakers & really the only optrion we see is to get another one & transplant transformers & glass. The protection in an amp here was the cathode resistor & later amps have this as standard, but probably why these early ones are rare is they failed long ago. A question is: how would you know any transformers & output valves had been seriously overheated if it worked? It'll still have a lingering smell is how, but old amps smell of many other things, the burnt caramel-coffee smell of fried transformer windings and burnt solder flux does linger as the room smells of it. We'll try to get the glass printed as there are shop window companies to do this, but whether it'll be more than an ornament or ever trusted is a lot of use & testing away which explains why it's a write-off as uneconomical to repair or possibly never be reliably safe for a non-tech person to use. To try to get it steady & reliable for the severe overheating may be possible, it would take months of testing & it might be safe again, but it might be on the edge of failing also. Would you risk a possible 400v on the speakers is the gamble. The broken glass is difficult as only shop window makers who bake the paint on like the original way, though printing on perspex might be a better solution. Perhaps it’s best back here after that but a bit sad looking, it’ll at least get a chance. So far, it lives, a huge amount of rebuilding with some further redesign such as adjustable bias & anything that looked tired. It appears to be Quite Famous now by web presence which means the odds of anyone finding one are zero now. But the work to get it right is extreme, we've looked at photos of ours as original & the work needed is daunting... Several months later it sounds heavenly if was very poor grade & took time to 'settle'. Made a new Tuner window artwork from the bits & got it printed on perspex so it has it's face again. For the upgrades & DC bias adjust pots to perfect it it's better than we've ever hoped for now. It stayed a while, but now has a new owner who certainly appreciates it. will we buy another one, unlikely, the work involved to buy as a resell is way too much. Anyone want one upgraded or actually rebuilt will need deep pockets...
What else is there in Vintage Valves? ↑
The 'Other Amps' page gets many valve amps we see that are interesting looked at for how good the circuitry is & whether it's worth upgrading to the extent the Trio got. The Trio is easily the best looking Vintage Receiver & that's why we carried on with it for so long. But now looking farther afield into the USA ebay listings which we've not done for quite a while, there are a few premium Brands like Fisher, Scott, Marantz, Dynaco & Harmon-Kardon as well as McIntosh that we are aware of. Plenty of average 10w early 1960s ones that are just too early & not worth the effort. Using the Valve amps we've had so far, 1963 & when Diodes were used as Rectifiers is really the 'modern' vintage we can deal with. But all will have AC heaters on preamps, weak power supplies, risk of obsolete valves & all that nonsense we can't be bothered with as the ill-fated Sansui 500A proved. To try any of the preamp & power amp(s) ones won't happen, we'll stick to the Integrated ones & are currently interested in an Amplifier or a FM Stereo Receiver. There are kit amps as with UK's Heathkit, so to avoid Dynakit & Scottkit as these are only going to be basic ones, if nice lookers.another problem is many of these are without the wood or metal cases, as chassis models made for console building were very common in the pre 1968 era. A case is a must as well as having one not rusty or weather beaten dirty as some look. Are we talking ourselves out of this perhaps? The USA ones are often $600-$1000 for a used, possibly working but still in need of a full rebuild & at that price it's not economical to import one with high shipping & customs charges. We've had the McIntosh valve preamps years ago & they were nice indeed but even then they had too many adjust pots & controls as well as now seeing how much needs rebuilding, the cost is too high & once selling it on, would the buyers like such rebuilding at the high prices? Probably not. If these USA ones turned up in the UK it'd be worth a try, but they rarely do. Looking on European sites, very little vintage valve gear beyond the low end ones. A case of see what turns up that is at least 20w & avoid the USA ones which is a shame as there is a big selection. But for the weight & import charges adding £200-£250 it's priced out. The Cold Hard Facts are still the vast majority of Vintage Valve Amps are just Not Very Good for upgrading to the Standards we can get from the Best of the Transistor ones. The Trio WX-400U was perhaps the best beyond the McIntosh-Fisher-Marantz-Scott-Harman Kardon-Sherwood ones that usually didn't get imported to the UK.
What An Ideal Valve amp Would Have... ↑
The way to deal with this is narrow down what we'd require in a Valve Amp. We know valve amps very well so can give a list of what we like. Next as we are in a surreal fantasy land of pure imagination we'll do a section on the Ideal Woman and how to write the Perfect Song...
1. DC regulated heaters on all but output valves.
2. Output valves should be something more substantial like an EL34 or KT66 variety.
3. Power should be 15w or more.
4. Something that looks good & is made to free-stand & be easy to use as such.
5. Proper Tone controls & Mono switch.
6. No useless Filters or Loudness Control.
7. Uses readily available ECC81-82-83 preamp valves or can easily be converted.
8. All-valve Phono stage.
9. Individual Biasing for each output valve & not fixed or cathode biasing.
10. Has a quality splitter circuit
11. Has a Headphone socket & associated load resistor circuit if the circuit needs it.
12. Is neatly hardwired or mostly on a PCB.
13. Has a mains choke.
14. Has enough space to upgrade to be the best spec & still stay tidy looking.
15. Has a clear circuit diagram available.
16. Doesn't smell of age or overuse.
17. Is actually findable...
You may think such an amp does not exist, and it seems this is the truth, or will be megabucks to get most of these fatures, after all no modern valve amps have Tone Controls. But we've been looking at the 1979-80 Luxman & Radford early revival amps, so go see the Amp Reviews page as the Luxman LX33 fits the bill, but has lots to upgrade. They will have sold very few as 1979-80 was the Receiver Wars years, valves were deeply out of date & even Lowther gave up putting their 1959 designed unsold stock in the HFYB by 1976. The first revival of valves faltered with these 1979-80 ones until by the mid 1990s the Hifi mags were getting interested again, with the Tube Technology gear, the strange Tim DeP Rogers valve amp & then the glut of cheap Chinese amps you see with ridiculous sales techniques on ebay & other sites today. But for a starter amp, many will try the Rogers Cadet III that has some of the features we desire, ie 4, 5, 8 & 11. Even valve amps that aren't very good will reveal a sound that you will forever be wanting. But be sure even the 1979-80 ones will need a lot of upgrading to be their best, after all the ideals of sound in 1979 were very different to the Magic Years of 1967-69 when progress in Transistors brought designs that were the best ever made. The revived Luxman LX38 Signature has some of this, but the transformers are still lesser quality with only 3 wires not 5 so not Ultralinear. To get all of these things, the only way is to build it yourself unless you dare to get an amp of good looks & rebuild it severely.
As a closing comment, just for those who read down this far. Valves will better any transistor amp, if a small few can nearly do as well, but the trouble is many valve amps are just rubbish designs. There are many 1980-today that just lazily rehash the old circuits & one popular one has an utterly awful stage in it that limits the fidelity hugely. But it gets reused still today as it's a cost cutter & as nobody really knows what Valves sound like done properly there will be many praising these poor designs as they will better most transistor amps post 1980, but it's only a glimpse of the real valve sound. But we do as we've taken 4 decent valve amps & just about rebuilt them, some to the point of nothing beyond fittings remain. For what that sounds like, try the 'Trip' section above. Nothing sold in shops is ever anywhere what we've made some valve amps into. But for those who are chance-takers in hifi design as we are, the clues are scattered amid the full range of valve amps if never all together. Your Assignment, if you Choose To Accept It, is to find the perfect mix. Your Reward will be worth it...
Why we dislike Amplifier ICs & why they are Never 'Hifi'. ↑
We had this page online for a while until late 2013, but now revived & updated, so it'll be updated, as we Still Don't Consider An IC has Any Place In Hifi. Some of it is still 2012 & it may read a little basic now but it stays.
There is more on this on the Deciding Top Amps page.
Why your modern "Hi Fi" will only ever sound boring & average before you've ever tried it. ICs in the Audio Stages. Assuming you even know better Hifi to realise it's mediocrity, most people don't understand audio quality or what they are missing so it's why it's still sold as average & ICs suit averageness so well. Audio today is not considered anywhere as important as TV HD picture quality, but Sound gives so much more that it needs proper amplifiers & speakers instead of the Puny TV Speakers as were complained of in the 1950s, nothing much has changed on TV sets in terms of sound. It's up to you to upgrade. ICs certainly have their place, and have made electronics a better thing in a huge way, but in Amplifier Stages they are a huge compromise to us. We are talking about ICs in amps we know from 1969 to the 1992 Marantz PM62 & beyond with the 2007 Marantz PM6002. We bought a 2002 Arcam FMJ31 preamp when it was new, it was all surface mount tiny components & ICs yet the Phono stage was older type parts but with an IC still.
An IC is just a general purpose design ↑
It will do the job safely & blandly in a range of uses, but clearly adds distortion through such generalness, all without any quality a designer of a discrete traditional transistors & resistors circuit can do so much better. A transistor is a Class A device unless used in another way. An IC has to run cool & be small in Hifi uses so the Class B push-pull circuit you can see on IC spec pages that show the circuit equivalent inside the IC & adds more problems than it solves. An IC will never be designed to get the full high power-resolved signal from an audio input, it can only create a "soft" general-purpose idea of it, for cheap hifi to the undiscerning masses, they know no better. In Tuners the IC was used from the late 1960s & to find a tuner with no ICs is pretty unusual, our Sony STR-6120 tuner has no ICs and sounds very sweet & natural for it. Amusingly Marantz must have read this page early on as they say the same thing in their advertising for the PM6004 ranges. ICs do not allow the Designer to put any improvement on a basic sealed-box IC. So we influenced Marantz...
ICs are Used Because 90% of Buyers Can't Tell Better ↑
Since first writing this page, as our 'Other Amps' page shows, we've looked at many amps & been Quite Disgusted At Them. Quad forever used to tell us their Audio Gear "Is The Closet Approach To The Original Sound". But their Quad 33/303 & 44/404 are mediocre items in our view, the 33/303 we give a honest view on it's averageness & Quad knowing that this junk was accepted so readily thought to stuff the 44/404 with lots of ICs as buyers can't tell. Cynical, but they were right. There's the thing: most people Just Don't Know the Better sound of the 1965-67 Transistor amps or Valves done properly. One reader inspired us to revive this page as it wasn't clear to him why he couldn't Hear The ICs. The answer is comparable to the Wine Buyer, a £3 bottle of plonk or a £1000 bottle of a perfect Vintage. 'Pearls Before Swine' is also comparable, you only understand what is better by having tried better, but the Swine are too ignorant to know better. You, as the Reader of this realise there is better than Mass Market Hifi Fodder and want enlightenment. This naturally goes hand in hand with Money many will think. The 'High End' market means High Price with £10,000+ being spent on one hifi item.
How To Justify An IC Power Block Stage...? ↑
As edited from our Other amps page, this amp has the cheek to say ICs are better! Trio-Kenwood KR-7300 This caught our interest as the Sept 1976 Hifi News has their advert for it & the deluded justification for using IC Power Amp Blocks... "To eliminate unevenness in performance caused by lots of individual manufactured parts, we developed one integrated unit called a Darlington power block". Er, if you say so, you're the experts after all... will have been the unaware buyer being told an IC general purpose design is better than discrete components. This is an open lie, all transistors, resistors & capacitors have a tolerance level, transistors are graded sometimes to the HFE level so you can choose. But the NFB in the amp will balance even the most randomly picked components, so the lie can be proved. The circuit looks pretty decent, all transistors & not overdesigned, but the killer is that IC output block TA-80W no specs findable, but it appears to cover the splitter, driver & output stages as a 10 pin STK type block unit, one per channel. Be sure these are long obsolete, so the amp is unrepairable if one is damaged, unless you fancy building a proper power amp? Amps with proper transistor power amps you can forever repair, but IC block dead means no repair possible. Thankfully only cheap AV 7.1 type amps still use IC power amps, the idea of transistors is thankfully the way again, but...
How To Justify An IC Preamp Tone Stage...? ↑
On looking at the 2010 Yamaha AS-500 85w £280 amp as on the Other Amps page, the Entire Preamp & Tone is on one IC with surface mount Ceramic capacitors. The power amp is all transistors & pretty much like a 1977 design. But considering the 1977 Yamaha CA-1010 was £400 in 1980 which is £2000 in 2010 money, they have to cost cut with these ICs as the tight-ass buyer wants far too much for their money. These sort of amps are ICs for Phono, Preamp & Tone pretty much as standard if the fad of IC Power Amps has gone back to transistors, probably as repair shops complained about the high replacement cost of one part instead of a 10p transistor.
What is Hearing an IC is in the Circuit? ↑
We should explain this further. In the early 1980s amps when ICs were used a lot, from budget to top range ones, the IC shows itself as adding a Grainy sound to the music. This is also equally due to Low Spec in the design that is typical of all hifi by 1979, even the best ones can be upgraded. Having upgraded the 1986 Realistic STA2280 as it was the first Amp we bought new, to recap it upgrading it brought about a big change, but ultimately it still sounded rough played louder & for 60w it was weak. There will still have been much low spec elsewhere, we done this in 2012. The Realistic had the hideous IMX stereo expander & Loudness, so teenaged us knowing no better used both to try to get Bass out of the thing. On trying it in 2012 like that it sounded ridiculous & still unable to give real bass as the circuit was so limited. Moving onto late 2014, we upgraded the 2007 Marantz PM6002 similarly, this time going to extremes by sorting out every spoiler & weakness we could find. The amp came to live & sounded a lot more natural, but we could "hear" the IC that was in the preamp & tone. The PM6004 mentioned below claims to have no ICs, yet they forget there are actually FOUR Stereo ICs in the Tone Stage. The listener will need a trained ear to hear the 'effect' of an IC in a modern circuit, they are fine-tuned with external components & perhaps better understood so they don't sound rough anymore. But the telling aspect is a certain flatness to the sound, a bottle-neck where the sound degrades. Only by playing many earlier valve & transistoir amps, especially the 1965-67 ones will the fresher more dynamic sound be known. Here we tried everything possible to better it, 120 changes is excessive. But it was never possible to get the fresh crisp open treble from this amp, though there will have been further changes possible, these wouldn't better the treble from how it sounded. The trouble with ICs are they are just General Purpose circuits designed to do the job 'good enough' which is generally what the market accepts knowing no better. ICs of the 1980s are probably still used if with better external designs, the early ones were very rough sounding, the 2007 Marantz one was still smooth even with most of the 120 upgrades done.
ICs to be fair are better used today ↑
The advances made with design in VHS, DVD & Computers have made IC designs much better, if only after years of not doing it very well yet having you believe it was better than what came before. In 1980 when ICs in Audio took over even the best brands, all they had were Sinclair & Spectrum computers & Sony using an IC in the 1974 TA-1150 so badly they never bothered to use matched ± voltages to supply it, look at ones today, they use properly regulated supplies with an idea that they know a lot more than you're being given. Why does that amplifier need an xxx in the yyy circuit? Usually just to make sure it's totally idiot-proof. Sod the sound quality: "we don't want complaints" is the thinking today & with how PC & Health & Safety riddled life is, only nothing short of calamity will revise the World's thinking... Yet despite this, every technology item you use, you think "What Idiot Designed That?" as you encounter poorly thought out menus & controls. But as you don't complain about that sort of poor design, only the not-working-right-at-all ones, the makers think it's OK. But they should get their products used by 100 aware-of-tech people of the widest differences & get their opinions. But no, that would cost money & also giving All The Good Stuff At Once like the "Back To The Future" joke scene would mean they'd have no Selling Strategy as you'd not want anything better. So no, you don't get it all at once, forever "improving" after getting it right 50 years ago & no chance of getting 10 years use with 2-5 years use the lot these days.
Suggestions For Improvement: If Hifi designers are really as good as they think they are, we suspect they are as frustrated using stock ICs, then why don't they design the circuit as they want, then put it into an IC & keep it just for that amplifier. The 1971 Hitachi IA-1000 & SR-1100 used a Hitachi made basic IC, you can tell there are transistors inside it even, but it's their design so used & made for a particular use. If Hifi must use ICs, why not custom make them instead of just using the same ones? ICs today are often surface mount, so the old op-amp chip, inline ones or 'Millennium Bug' type ones are not used as too big. Why not design the circuit to perfection, stop putting the stupid excessive spoilers in & give Hifi buyers something Decent? But that will never happen in this Corporate world, they want you to buy a new one often & couldn't care less if you are finding fault with the current one after 2 years as a new one is out & the sheep readily buy it, the i-Phone is the best example ever of consumer gullibility. They don't even keep spares more than 5 years, with the remainder ones just disposed of as rubbish or recycled, not sold to those who would sell them as spares. Cynical Money-Grabbing Lying Smooth-Talking Corporate Bastards is fair comment to all of Consumerism today. But to be two-sided here, the Consumer is to blame forever wanting it Cheaper & Cheaper, yet still thinking Quality can be given. The Price is all that Matters today. If you want to buy any item of Better Quality than is the usual fare, going strictly by price, beware you are likely just overpaying for the same crap. There are sellers that sell Amazon's stock on ebay for £1.50 profit, to see goods supposedly better quality as priced higher is usually ebay sellers pretending they were their items & it arrives in an Amazon box from Amazon. There is generally no better quality available beyond buying a Vintage one.
Not All ICs are actually "Real" ICs ↑
The most common one is the Double or Triple Transistor, also a Double FET. Some early ones are called "Thick Film ICs". These are not really ICs as they are just 2 or 3 units in one package, they are Not Integrated Together inside. You see these on amps from 1975 onwards, the earliest one we've seen is the Double FET "IC" on the Sony TA-3650 with 2SK58 giving the FET numbering away, also the µPA68H is another common one. These are not to be feared, once identifying them as not being similar 7 pin amplifier ICs or ones that look like op-amps as the 2SK58 does. Again these are used to save money, but for the user they can be a pain as they are often obsolete & expensive as well as just one part failing needs the whole item replacing.
It's Cheaper To Make with ICs ↑
But back to ICs. They are simply used... To Save Money... as the Buyer Won't Know Any Better. Why should an item have a complex preamp designed when one IC & a few components can do the job nearly as good. This is the point, 'Nearly As Good' to which we agree. An IC stage will still deliver the sound & 90% of non Hifi Enthusiasts will be happy with it. But those who know better Hifi can pick out the 'averageness' of the IC. It has no huge dynamics, no wide Stereo soundstage, no deep Bass fundamentals, no Smoothness to the sound & therefore Treble is poorly defined. The designer knows the amp is Crap but they just then dumb it down to sound acceptable to the Mass Market buyer who won't understand the deception and will likely be happy with the Audio Item until it packs up after 2-5 years & then they'll buy another one. The New One will always "Be Improved" making out manufacturers care. But having tried the 2007 Marantz PM6002 we can see the design is no different to the better designs of 1977 such as Yamaha were using. The "Improvements" were using very fine board track that you can't really resolder as it'll pull away & using tiny 1/8w resistors, if not quite going to Surface Mount. The 2002 Musical Fidelity A308CR pre-power amp we unwisely bought ex-demo was exactly this too: An Insult. Lots of money spent on huge casework, but inside just average TV grade components & lots of cheap ceramic capacitors instead of anything Quality.
There is the risk of some ICs in the Audio stages ↑
Any amp from 1969 onwards can have ICs & on researching ones we'd like to try to upgrade for our research, we are careful to avoid ICs. In FM-Tuner stages it's more acceptable as at a low level, if still noticeable to Tuner enthusiasts. But in Tone-Flat Amp stages & even pre Power Amp & ridiculously Output Stages on some supposedly premium amps, buying these IC riddled amps means they'll only ever sound rough. Many buyers at the time will never have known there was a problem. Only by seeing the amp itself, or more simply, if available, get the Service Manual & see the Circuit Diagram. Look for big triangles as they mean ICs, but be aware some Block Diagrams use the symbol to denote a section, not it being an IC. An IC is the familiar black flat rectangle with multiple legs, or it can be an upright comb style black piece with 7 legs or more. Early ICs are like the Can transistors, but having more than 3 legs shows they are not a Transistor. But early Germanium transistors had 4 legs, but you'd not expect an IC pre 1968 anyway.
We've Upgraded many all-Transistor Amplifiers now ↑
We do this often just to see how good they can be made & done similar with IC heavy designs. The first one we tried in 2012 was the Realistic STA2280 which was still deeply limted by a weak power supply, lots of ICs in the audio & a general cheap construction. More recently we tried the 2007 Marantz PM6002 45w amplifier. It needed 110 alterations to remove the huge amount of dumbing down. It actually sounded quite decent, but further listening & compares revealed the one Pre-Amp IC & the two in the Tone still did affect things. It was still low spec sound despite the Lies Quoted saying "How Improved" it was. The all-Transistor amp, if a quality amp to start with can deliver sound so accurate & focussed it blows your mind, yet the IC heavy design will only sound of it's weakest link: IC op amps with 12-17 separate mediocre designed transistor stages within creating stages described below.
Is New Hifi "Improved" From Older Designs? ↑
No. Just No. Hype in Selling New Hifi Amplifiers goes on in a sickly way saying things like "Our Designers have Toiled Hard To Make Our New Amplifier Better Than Before" and "New Improved Wash Whiter Than White" and "Mk II SE version". All this is utter nonsense, it makes you feel cheated "They Sold You Rubbish Before" but no-one ever complains about that. The "Advances In Technology" cover this. With our Hifi testing, upgrading & reviewing, we can tell you about one of the Best Ever Transistor Amps: It has ample power, 50w, it sounds fantastic with a real open sound with correct bass & a punchy sound. It has a very low noise floor, on Aux the specs are 90dB & today 100dB is quoted as being of very high quality. The noise level of 90db means it is 90dB below the full 0dB level. For those into design, it has buffer stages & correctly applied NFB in separate stages. It has high current capability like modern amps as it has Doubled Output transistors. It has full Tone Stages with Defeat switch & a MM phono stage. It has Pre Out-Power In connectors too. But this amp is extremely rare & the manufacturers realised their mistake in selling Such Perfection as the 'Mk II' version was released 17 months later & the first version was not advertised or widely sold as only 5000 were made & be sure few exist today. Yes, it's the 1965 Sony TA-1120 amplifier. What does this Amp lack compared to Modern Hifi Wonder? Er... nothing beyond MC Phono. The sound of it is excellent. The only thing Obsolete on the design is it is Capacitor Coupled, it's not Complimentary though it uses NPN & PNP driver transistors. It has no need for Differentials.
There has been progress in Amplifiers as in losing the Capacitor Coupling, adding Differentials & going from Semi Complimentary to Fully Complimentary. But This Isn't New Either. The 1967 Sansui 3000A & 1967 National Panasonic SA-65 receivers, both 45w are actually Semi Complimentary using ± HT voltages, not the single 90v maximum the TA-1120 uses. The only reason they are still only 'Semi' is the high cost of PNP output transistors & for a 45w amp this wasn't really possible until the early 1970s with 1973 being the first year we note Fully Complimentary.
Progress from here on is non-existent in reality. Only to cost cut. Only to put it into ICs. Only to add on extra pointless stages to cover up failings by one mediocre IC stage to still get the coveted but worthless THD & IM tests. The Unrealistic High Specs of 0.001% THD are utter nonsense, as from our Power Ratings page...
THD-Harmonic Distortion: In Your Face. ↑
To put it bluntly, the reliance on THD in maker's specs is utterly worthless BS. It means Nothing in terms of how the amp will sound. Harmonic Distortion is worked out using a Sine Wave & reading how it distorts on it's Harmonics, read elsewhere about good & bad harmonics. So the 1kHz test tone gets 0.01% THD, how fun for those who listen to test tones. Music is a multitude of frequencies all at once, it relies on the speed & reserve of the design to reproduce it. Square Wave distortion is another one quoted in reviews, slew rate & ringing. All amps rely on a design based on mix of a large quantity of transistors, resistors, capacitors & diodes. The designer may know how to fully bring the best out of a circuit, but then comes the brutal dumbing down so you don't get the Good Stuff as you'd never buy another amp. The only true test for Distortion is how the Source compares to the Output, it will usually be tamed down to not be so dynamic, deep bass will be reduced, treble can be blurry or smeary. With today's technology, to sample 5 seconds of music at the Source & at the Amplifier Output, scale them to the same visual size & see where it goes wrong. That peak got reduced by 10db, that transient didn't resolve correctly, the Sub Bass has gone missing. The "Distortion" could be of several Hundred Percent where it fails to resolve complex signals of Music, not easy Sine Waves. Go on Hifi "experts" put that idea into your sales hype & tell the truth how mangled your crappy amps are. Then stop dumbing them down so heavily like 120+ changes were needed to better, if not possibly perfect, the 2007 Marantz PM6002. Hifi should be much better than it is.
An Example: 1992 Marantz PM62
The Service Manual describes the stages in the Stereo preamp AN7062 IC which is interesting. For a start one IC for 2 channels is a cheapout. It shows the IC contains a large amount of transistor circuits, the Yamaha CR-820 preamp uses a mere 5 transistors in the preamp per channel, how many does this contain? Differential Amplifier, Current Mirror, Cascade Bootstrap, 2 more shared Current Mirrors and an Emitter Follower aka a Buffer stage. The pins mention NFB, Linear Feedback & Phase Compensation. Sadly that's all they'll let you see, no equivalent circuit. So go through them one by one. A differential amplifier is a type of electronic amplifier that amplifies the difference between two voltages but does not amplify the particular voltages. It's the ± input one with a common output. It's a Long Tailed pair as seen often in mid 1970s amplifiers. The Current Mirror comes into the circuit here too, we can't really see the point of any of it really when an ECC83 valve just adds gain & does it so sweetly. It sounds like problems being created by losing gain & solutions added to sort out what you just messed up but why? Buffer - Emitter Follower circuits add no gain, if actually losing a tiny bit, but it's a matching & impedance thing. Cascade Bootstrap sounds like a sampling thing for impedance variations, ie to tidy up the mess the other crap does. Transistor counts based on sample circuits: DA=2, CM=2, EF=2, CB probably 2. So together with the 2 extra CM the IC contains 12-25 transistors of over complex BS nature that seem to make a mess & then remedy it which is why the sound is poor. Illogical. But it will have resistors & capacitors too, probably another 12-30 at least based on the Phono one below. Keep circuits simple is what these stupid IC designers have little idea of. It all sounds like rubbish to us & sounds it too. The phono circuit is what we see now as the Differential Amp into the ICs for MM & MC. The IC is an NJM2068DD, at least not an LMxxx but the same type. Burr-Brown is the "premium" IC type mentioned earlier, a "precision" item that still costs 40p. It's like grading Turds... The NJM2068DD actually gives the component equivalent if not the values, the symbols. Looks like a DA & CM again. Per channel it's got 15 transistors, 2 diodes, 9 resistors and 2 capacitors. All in that tiny thing. No wonder it sounds crap. A typical early 1970s Phono stage uses 3 transistors per side, some use as little as 1 or 2 on the 1960s early designs, a Sansui hybrid 1966 amp has just 1 per side.
Marantz PM6004 shows we've been Read. ↑
This is the write-up Marantz have on their website for the PM6004. It takes the Idea as we wrote above that ICs do not allow the Designer to put any improvement on a basic sealed-box IC, as an IC is just a General Purpose item. The amp is now obsolete, as in model upgraded, but we'll give our opinion on the nonsense they write, to read this it sounds very exciting, but how Real is it? Are We Being Lied To? It would be too easy & beneath us to punctuate the below with comments, so we'll refrain from doing so...
When “What Hi-Fi Sound and Vision?” gave our PM6003 the “Best amplifier in its class” award, the performance bar was well and truly raised for its successor: the PM6004. Fortunately, Marantz design engineers enjoy nothing better than a real challenge. And they seized this upgrade opportunity to further improve this already excellent amplifier by refining the sound and optimising the balance from Bass to Mid to Highs for a performance full of harmony. The strategic difference is that now both the pre-amplification and power amplification sections exclusively feature discrete components – not a single IC in sight. Specially selected single components not only outperform ICs in quality, they also make it possible to refine the sound in much more detail, resulting in more clarity and transparency in the sound stage. Of course the lower impedance Toroidal transformer with case-shield has been retained, so too has the customized main capacitors, and the potential power transistor which is mounted on the extruded aluminium heat sink and can easily handle higher currents to keep close control of speakers. All of this proven, superior technology has been retained. And of course all of the original features - such as six-line audio input including phono, two-pairs of gold-plated speaker connections with the convenience of bi-wiring, and a remote handset - are still there. All that’s now required is a lover of music.
Translated From Hifi-Bee-Ess language into What You'll Understand: Well the idea is the 2009 PM6003 is the same amp really folks, the one they're clearly saying is rubbish now but they raved about it a year before, yet the 2011 PM6004 adds the "new idea" of transistors not ICs, like it's a Radical Concept. But the PM6005 is no different apart from adding a cheap DAC to the insides. You're only allowed a little improvement at once, all cosily worded to give the idea it's important & why you should have it, sucker. The Weasel Words & insinuation that this is the Best is sickly. The idea they all sit around Playing Bears all day & decide to do something just for you is laughable. The PM6002 is little different to the PM6004 as we circuit compared. In fact, their selling hype is a LIE: they claim to have no ICs in the Preamp, yet they forget there are actually FOUR Stereo ICs in the Tone Stage which is a Preamp stage as every other amp reveals. The ideas of Amp design are so close to what Yamaha puts in their 1977 amps. There is nothing new. Some of the sickly wording borders on idiocy, giving the unaware the idea a lot of this is New Thinking, whilst others are made weary by such cheek. The wording clearly implies all their IC based Hifi for the last 30+ years is inferior, so start queuing for Your Refund Folks. Making out ordinary things are important: a toroidal transformer is cheaper to make & the case covers up an ugly finish. The importance of the Heatsinks is nonsense. Things Proven Superior, yes, by Marantz & others designing Transistor Hifi since the 1960s, but oddly they are always 'improved' now. The scam of Bi-Wiring gives no improvement except to cable makers who sell double. Good things from previous versions are still there, but you can't have all the "improvements" at once, you have to wait for the PM6005 to get a cheap DAC built in. Saying 'What Hifi' by virtue of paid advert-reviews is no recommendation. Specially selected Components is amusing: ones the cheapest to buy in bulk that just give the pointless THD/IM specs but little more. You may think that they may have Better Transistors with Lower Noise levels by now, compared to 1960s ones, but on upgrading our PM6002, the noise level was no different: the huge amount of spoilers lessens the background noise. They are no better at all. The Power Capacitors in the 2007 PM6002 are only small 10000µf ones, the modest 1978 Luxman R-1040 uses the same value. The main improvements in Modern Hifi is the User Control with techniques learnt more with VCRs & DVD players to give useful Remote Control. You can read our Review of the PM6002 on the Vintage Reviews page so we'll not repeat more than saying that EVERY opportunity to dumb down or limit the sound was taken, It took 120 alterations, removing useless items, losing limitations as well as upgrading. This is Dishonest Hifi & even with the 120 upgrades the PM6002 still was limited by the IC hidden on the main board by the cable socket from the Volume, it looks like it's a Control IC but it's not. The weak contruction & too thin board track just about stops upgrading & certainly stops experimenting. These are just general items destined to be E-Waste as they are not repairable to a good standard like the ones even pre mid 1990s. But the funniest thing of this is... go read the 2009 PM6003 write-up, it's almost the same. Now why is that...?
Not all IC amps are Bad, if they'll Never be Excellent. ↑
Just goes to show exceptions are possible. The Teac & the Photax-Concertone have ICs in their design. Both are from 1971 & the Photax arrived the same day another Teac did as we wanted to max it out as the sound was something special. But the Photax got the interest as it sounds remarkable. Later ICs can contain many components, these 1971 amps have very basic ICs, the Photax has 3 transistors & 5 resistors per IC. Perhaps the problem with ICs is they got too complex & designers don't know how to design either ICs or the rest of amplifiers to sound good like they used to. Buyer interest in hearing high quality Music is very low despite HD Television. But the Teac & Photax with ICs are not cheap mass market amps, the Teac is genuinely rare, the Photax is totally unknown except by us. The Photax we can find the inside circuit equivalent, but the Teac remains unknown, though we'll search again. To damn our overall view of ICs as being "crap" for audio, the Photax has the best sounding Phono stage we've heard. Only because it's a basic one, your 8 pin op-amp is a very different beast, see the Bad ICs section below. We've maxed out the Realistic receiver to see how it coped with all it's 1987 ICs & we improved the sound a lot, but the ICs were still very obvious, read the Realistic STA2280 review.
No escape from ICs in Modern Digital Audio ↑
We use a Computer Sound Card to output sound from the Computer to Vintage Amplifiers. It uses an IC based Soundcard. We use a TiVo box for TV sound & the sound goes into a cheap DAC that cost £10 if we upgraded it inside. It uses an IC. So there is No Escape from ICs in Audio. These can deliver Audio to 0dB level so as it's Line Level there will be some compromise. But with Digital TV being your only option today, there is no Valve or Transistor equivalent. The only way to get Audio as Analog & with valves or transistors is to play Recorded Music from Vinyl or Tape. CDs are Digital & involve ICs. The Music We Play is recorded from Vinyl but Digitally Sampled but is mastered to Our Sound & we don't like how commercial CDs are recorded & certainly can't listen to MP3 upsampled to CD quality, the high treble is just not complete enough, though you could process it, digitally, adding Valve harmonics to smooth the sound. We have therefore succumbed to ICs whether we like it or not. At the Source stage we have no real choice. But when it comes to Amplifiers, the 0dB line level signal is reduced by the Volume control then it goes through various amplifying stages to the Power Amps. Any IC based amp with only ICs, no discrete transistors will still deliver a sound good enough to please the Masses who know no better. But the IC just being General Purpose & often bought off-the-peg from another manufacturer is the problem. IC data sheets give suggested circuits, be sure these are averaged & safe for all use. Hifi playing safe is Boring Hifi. It will give good THD & IM ratings (Total Harmonic Distortion & Intermodulation Distortion) but it can be made very low spec & the buyer will be None The Wiser.
Are Modern ICs Better Than 1970s-1980s ones? ↑
The progress made with CD players when at one time there were several competing types of sampling, bitstream PCM etc shows progress was made. You hear nothing of that since the 1990s as things have settled down. Early CD players like the 1992 Marantz CD52 Mk II SE don't sound very good now, similarly early cartridges like the Garrard G800 now sound so crude. But a CD player for the Computer is now a DVD burner too & OEM ones were £20 last time we bought. ICs will have certainly improved, but the General Purpose Average Design of them will still be the issue. ICs have to be designed to run as cool as possible, the amount inside a Quad-Core Computer Processor means it runs extra hot & if it had no heatsink it'd not last 5 minutes. So it can be said that Line Level Audio ICs are better, the 2007 Marantz PM6002 did sound better than we'd expect, if only after 110 issues were sorted. But the board track was too weak to have it be reliable, so to try to upgrade later amps can end up with issues.
Source Direct Is A Total Lie. ↑
In IC era Amplifiers, Source Direct is often touted as Important, such as CD-Direct. So you think it means CD plugged in goes straight to the Power amp without anything else hindering it beyond the Volume Control? You Have Been Conned. Early amps such as the 1965 Sony TA-1120 used the idea of Tone Cancel & the 1975 Yamaha CA-800 II has the same idea. But the Audio still goes through amplifying stages, it just means the Tone Controls are bypassed if their circuits are generally still used, it makes A-B comparing easy is the idea, or to leave the Tone set for one input but set it flat for another. Some like the 2007 Marantz PM6002 use 'Source Direct' as did the 1992 Marantz PM62, but closer looking at the PM62 shows the signal still sees all the amplification stages if with a fixed resistor to keep the Gain the same, as do the 1965 & 1975 ones. The 2007 Marantz goes a bit further, it bypasses the Tone & Balance stages but still sees the preamp IC & still has resistors fixing the gain. On us upgrading the PM6002 we bettered it to the point 'Source Direct' sounded less good than our upgraded version. If the Source Direct button done what You Think It Does, the gain would be much too low as it'd only be using the Power Amp for gain. Similar to those using 1957 Quad II power amps without the Preamp: the amps need 1.4v for full 15w, but using the Preamp it's rated at 70mV RMS. Therefore those using the Quad II with just a modern preamp won't have enough gain, the Quad II has no preamp gain valve, it's in the preamp. This explains the volume differences. There are power amps with very different 'input voltage for max gain' ratings & if yours has ample voltage gain with extra preamp stages, you can go waste your money on a Passive Preamp, which is just an input selector & volume control, and feel it's "so much better" without the active (amplifying) preamp. You are being lied to again.
Those Nasty "STK" Block Amplifiers. ↑
In power amplifier stages they are the Ultimate Sin as the ultimate cheap out. But heavy use since the 1979-82 era once cost-cutting got heavier, after being introduced in the late 1960s with the Rogers New Cadet having a very early STK block one. They are cheap to make & use circuits around as the IC lump thing contains all the parts that make a design individual. But that's the problem, the lump of the IC amp is all there is. You can't add tweaks or design specialties to it. They sound rough & once their max "clean" voltage output level has been reached, the sound fades into a nasty haze. Underpowered parts & the fact these can run Cooking Dinner hot shows something is wrong. One in a 1985 Sony we tried to get a better sound of, but any increase in gain was met with a huge gain in heat. Also if one part breaks down, the whole thing has to be replaced, try unsoldering one of those with 20 pins. Beware the lower end Silver Pioneer use these awful STK block amps & we're shocked that they creep into what is supposed to be top Hifi with 200w amp blocks in a Trio-Kenwood "monster" amp. A Sony that Forum types rate highly we found had IC op-amps & then 2 big IC blocks as the output stage. Crap. IC Block Package Amps are like Package Holidays, mass market mediocrity for the undiscerning, which covers about 80% of buyers, right? Only a discrete design meaning inddividual components chosen by a designer to suit a particular use are the recipe for the best sound.
No ICs Allowed In Preamp or Phono Stages ↑
This is our idea, but many manufacturers still cheap out & to find a 1980s amp without a Phono IC is almot impossible. These are the parts of the amplifier that give the amp it's quality or not. Tone controls are involved & if designed well with disrete components, the sound can be excellent. Phono stages at low level may benefit from the Low noise, but transistors designed in the late 1960s are still good enough so it's all down to cost. But with an IC you just get an industrial non-hifi IC part that cannot be given any class. Many amps use op-amps, 8 pin square things, we used to find it amusing rewading of fools thinking a better quality one was making their amp better. Again the rough edgy uncontrolled sound is the sound you get, from enduring awful 1980s amps after how great earlier ones are, the grainy bright thin bass sound really was what the 1980s sound was about. No sweet crisp treble or deep bass as 1970s amps trying for the valve sound offered. A lot of this is down to low spec beyond the ICs, but the IC is the weakness & you can't upgrade beyond the IC. Not even with Burr-Brown "Audophile" op-amps, nonsense, it's still an IC. Instead a teeth on edge nasty sound with Zero Fun Rating. This sound is with amps with or without ICs, but having maxed out the 1987 Realistic, it really did have shades of 1970s as we purposely upgraded it to sound like that, if to our 2012 ideals, but then upping the volume it just gave it's IC roots away despite a proper transistor output stage. The weak power supply will have caused a lot of the problem, but the flatness of an IC we can easily tell. ICs are used as voltage regulators which years ago we ripped out of our Valve amp preamp as hearing it with & without, the without was a huge improvement. The regulators were dealing with over 200v, terrible idea & a boring sound. Regulators are good for lower voltage & one in a valve heater circuit can be excellent.
Where You Find ICs in Amplifiers. ↑
If an IC is in the Power Output stage, the amp is doomed. From budget amps to 160w amps, using an IC here is unforgiveable. In the pre-power stage similarly, the IC is carrying higher than line-level signals & the mush it adds will kill the quality. ICs in the Flat & Tone sections are as bad too, the signal is at Line Level & bland IC design means it'll be mangled. ICs in the Moving Magnet MM Phono stage are usually nasty Op-Amps & are so low powered the chance of it resolving the audio signal beyond a gritty mush is never going to happen. ICs in a Moving Coil MC stage turn up in Yamaha amps, see the Yamaha page, but what they are is unknown as no data found.
We'd sell it as Used, so we Buy it as Used as one forum writer puts, nice line for letting someone else take the new-to-used price dip, but missing out on owning fresh goods. Sound is different to picture, sound is a fragile thing & delicate ambience that exists in older music can be harshly lost. Music today is rarely "live" like in days of old with a group or orchestra in a studio all playing at the same time & creating synergy by doing so. Today Auto-tuned vocals singing over a backing track is what there is & how it's all recorded & how disposable most of it is, ICs suit it.
Discrete components is the way Hifi should be made. ↑
The classic era of amps had those big TO3 oval metal transistors on big heatsinks. Did they sound good? Read our other pages & see. We are aware of many bad designs with no ICs but very few good ones pre 1992 with ICs. We have a Pioneer M90 + C90 200w pre-power combo here that we couldn't find the parts for to fix it, so here it sits a year later. The Preamp is stuffed with IC & the Phono stage has a huge amount of parts other earlier amps oddly never needed. 1980s amps usually sound pretty awful after how good the 1970s ones were, Yahama being a very obvious drop in quality from the 1978 to 1980 range, yet the 1985 A-720 is a step back up.
Do FETs belong in Audio...? ↑
(Updated Jan 2018 & on the Blog page too)... We thought FETs were the weakness in the Sony TA-1130 & other amps that used them in Audio Stages. The NAD 300 huge receiver had FETs in the preamp & we couldn't do much with the amp, but on later looking at the circuits, the whole amp is a bad design. Usually the FET was a bad idea trying to cover bad design as the Yamaha CR-1000 uses them on the preamp. But as with Differentials, we need to find out what is bad & what is good, no-one else tells you info on these things & for us to find out is the only way. Progress in Upgrading gets a 1971 Sony TA-2000F here, to see what a preamp with so many FETs like the TA-1130 has can be upgraded like. To leave the FETs & not try to replace with Silicon even based on the earlier TA-2000 preamp. FETs are in FM Front End sections of the Tuner & this was an idea first used in c.1968 with the Sony STR-6120. We've found Differentials are fine in Audio, but the trouble is, the Differential Era generally starting from 1971 is where cost cutting takes place, the Differential gives good specs if the sound can be awful. But it's not the Differential that's the problem. So to try upgrading the TA-2000F with all our ideas but leave the FETs be. It sounds as punchy, crisp & deep bass as a Silicon Transistor preamp. FETs may only have low spec in comparison, but a FET does not need much power to work unlike a regular silicon transistor. The TA-2000F FET noise levels are comparable to transistors, if not totally silent. Based on how they sound, the FET is actually 'no bother' as we found with the Differential. The big problem is Bad Design, Cost Cutting & Fear of The Real Sound to wrongly blame the FET in preamps. Is it better to have a FET or Transistor in a preamp based on Sound. Actually it's no different on a well designed & properly upgraded amp. The 'Fashion' aspect of adding the "F" was a Sony thing & they progressed to V-FET output stages a few years before MOSFETS arrived in 1978 as noted above. The only trouble with the TA-2000F FETs is they are awkwardly numbered & sorted into gain groups, if one fails or goes noisy, to replace is the difficulty.
Tuner Stage ICs ↑
Only ICs in Tuner stages at very low levels & in Moving Coil Phono initial stages are a bit more acceptable, if still a bit of a compromise but where do you go to escape them totally? FM with ICs in the later stages as the volume reaches line level is still as bad as a Tone section with ICs really. The Tuner in the 1968 Sony STR-6120 is based on the second version of the Sony ST-5000F tuner & it has no ICs at all. Once upgraded losing the ceramics, the sound is as sweet as a Valve tuner. The 1977 Yamaha CT-1010 is touted at the 2nd Best Yamaha tuner (after the CT-7000) yet for the huge case to match the CA-1010 amplifier, it had little space used & still has several ICs. The CT-7000 still has 7 ICs in the IF & discriminator stages. Why does it need all this when a Valve tuner & the no-ICs Sony STR-6120 can sound so good? The answer likely is looking for higher sensitivity & channel discrimination, but most people don't DX (search out distant stations) they use a Tuner for background music.
Moving Coil Cartridge Stage ↑
We've never cared to try a MC cartridge they were a 1980s trend for hair shirt wearers is our opinion, very high buy price & then to retip or repair is high maintenance. The MC cartridge output is much lower than a Moving Magnet MM usual one that poor Phono stages & added gain noise surely are worse than the potential top range ££££ amp benefits. Oddly as Phono stages are still around, any MC cartridges around? Tuner stages work similarly on tiny voltages, but ICs working on line level is where the roughness starts. A cheap amplifier can play simple low volume music quite well, but throw something complex & loud & the over-engineered IC takes over. But some of the Early MC stages still used Silicon transistors & the Trio-Kenwood KA-6000 is stated to have a very high noise level in use, as per the Jan 1970 Hifi news review, sort of negating the point of the "benefits" of the MC cartridge.
(early quite naive ideas... but it stays) In an ideal world, you could have adjustments to correctly balance every + and - waveform half to give perfect undistorted sound. ICs use this push-pull type design with differential amp stages with + and - inputs and a single output & if these are not well designed that's the rough IC distortion you are hearing & over all the badly designed transistors in the IC it just gets worse, this is why they sound so bad. There must be a better sort of IC that could be made that is of higher quality, but all they do is add extra stages trying to undo the mess they made before. See the PM62 preamp section above for more.
BAD ICs: What is inside them? ↑
LOOKING AT WHAT IS USED IN 1980 & 1984 YAMAHA AMPLIFIERS
older opinions... Clearly Yamaha wised up on the ICs as the 1985 Yamaha A-720 amplifier is all discrete transistors except the Phono stage. But the 1980 range was choked with ICs & their CR-2040 should be a great amp, but is spoiled by an IC Tone stage. The circuit shows Tone Control has ICs, one each channel for the main stage & another each for the Presence, losing the inductor of earlier amps. A Hitachi HA1457 IC is an 8 pin strip IC still findable on ebay as a HA1457W for £11-£17, are they the same? The equivalent circuit is findable for the HA1457 & it contains a ridiculous 15 transistor stages & 7 resistors, ICs suck (the life out of music) don't they. Those transistors are in circuit designs ie differential, mirror etc, but why do you need them? The amp will only sound rough on every source for using this complex but small IC on line level. There is no way to bypass it if you switch tone to flat either. Crap.
The 1985 Yamaha A-720 has firstly 2SK389 as the smaller pair, but just a dual FET, 2 FET transistors in one package like the 1978 amps used. We don't mind that. But there still are 2 more M5218L (or NJM4558S) and M5219L, in PNP and NPN type format. What are they. There are either found as inline or flat IC types with after letters differing. Oh dear, 2 channel ICs with again a ridiculous 14 transistors, 2 capacitors & 10 resistors & a diode in the usual configurations. All the usual Differential & Current Mirror things if done properly would no doubt sound good, but like this it's dismal. Odd how older amps done well enough with 3 transistors per side. The trouble here is the size & power of the circuit is limited, so transistors inside it will be designed "cool" as in valve terms & not be giving the full fidelity of the source it's true sound. A mediocre mass-market universal IC running cool also on Line Level is a rubbish design. You may think they'd have improved things by now...
Our 2012 opinion on Why ICs Sound Rough. ↑
We bought the same 2 amps we bought new back in the day. The 1987 Realistic STA2280 & the Marantz PM62. Both have multiple ICs. Both got treated, er mistreated at the time. Did we buy well? This was done in our earlier days of this current Hifi writing interest, which in real terms means we sold both finally in early 2012 after having them a few months. Neither are badly made in some ways with proper Transistor outputs, decent power supplies, regulated powr supplies, the Realistic ones ran typically too hot. Both had ICs for Phono & Tone-Line stage. The Marantz got only a few capacitors changed as in 1992 they used decent ones & values that are OK to us, the Bass Control capacitor was altered to give it a bit more richness & it didn't sound bad at all. The Realistic was a bit different, it used ordinary capacitors which were still in good condition but were very low values in many sections, so we recapped the lot except the 2 power supply caps. The sound overall was actually pretty decent but turn it louder & it just flatted out dismally as the regulators were similarly underspec. Ignoring that problem, we even lost all the Ceramics in the Realistic including loads in the Tuner stage. It certainly improved the sound & confirms ceramics are crap for Audio in just about every area. But then with the audio not restricted, at least playing within the power limits, both sounded a bit grainy & rough. A lack of focus that discrete circuits can have as the designer could use values suited to their spec, an IC is a lump that is mass-market & universal in it's blandness. And poor focussed power-weak gritty grainy miserable sound is all you can expect. One-size-fits-all may work in socks, but in Hifi there are Horses for Courses. In a preamp stage an IC is an insult compared to what an ECC83 valve or it's close transistor equivalent can do.
We have discovered far more with Hifi since with these two last sections, but this is left unedited as it still is worthy.
Modern Hifi: Why we Hate ICs. ↑
We consider an IC in an Audio Stage of an Amplifier to be a BAD idea. Phono Stage, Tone-Flat-Pre Stage & Power Amp Stage should have NO ICs. The only ones that are aceptable are the 1970s FETs with simply 2 FETS in one block & not really a circuit at all, like Yamaha & Sony use. ICs in Audio Stages is why your modern "Hi Fi" will sound mediocre before you've ever tried it. Don't get us wrong, we don't want a Valve Computer & ICs certainly have their place, and have made electronics a better thing in a huge way, but in Audio they are a no-no to us. We are talking about ICs in amps we know from 1969 to the 1992 Marantz PM62 & beyond, what they are like now is unknown to us, but we're not keen on knowing as it'll cost.
There is the risk of some ICs in the Audio stages ↑
on any amp from 1969 onwards. In FM-Tuner stages it's more acceptable as at a low level, but in Tone-Flat Amp stages & even pre Power Amp & ridiculously Output Stages on some supposedly premium amps, buying these IC riddled amps means they'll only ever sound rough. Many buyers at the time will never have known there was a problem. Only by seeing the amp itself, or more simply, if available, get the Service Manual & see the Circuit Diagram. Look for big triangles as they mean ICs, but be aware some Block Diagrams use the symbol to denote a section, not it being an IC. An IC is the familiar black flat rectangle with multiple legs, or it can be an upright comb style black piece with 7 legs or more. Early ICs are like the Can transistors, but having more than 3 legs shows they are not a Transistor. But early Germanium transistors had 4 legs, but you'd not expect an IC pre 1968 anyway.
We've maxed out an all-Transistor design ↑
to see how good it can be made & done similar with IC heavy designs. The all-Transistor amp, if a quality amp to start with can deliver sound so accurate & focussed it blows your mind, yet the IC heavy design will only sound of it's weakest link: IC op amps with 12-17 separate mediocre designed transistor stages within creating stages described below.
Early Basic ICs may be acceptable ↑
The Yamaha CA-500 & Hitachi SR-1100 use a ceramic looking block IC, not a typical looking IC. Having seen the equivalent circuit it has some fixed resistor values but the other ones are direct to the pins & you can alter & upgrade the design still. The Teac AS-100 uses a custom Tone Stage IC of unknown content, but if it compares to the 1969 AG-7000 Tone Stage it's just 2 transistors, but it remains a 'black box'. Certain very basic ICs are similarly just two transistors in one package. No other resistors or anything, you'll find Double Transistors as in Marantz 2265B tone section, Double FETs doing similar, these are used as the first stage in Power or Preamp stages as Differential Amps aka Long Tailed Pairs & are of no problem. Some ICs are other types of Double or Triple Transistor though some do contain Resistors making them more general purpose as the designer using them has to use fixed values. These are a bit of a compromise, but look at the circuit equivalents if you can find them. The RCA tuner in a transistor can with 8 legs type ICs as seen in many late 1960s Tuner stages are many transistors in one package, but a Tuner is not so important today. Those awful cheap op-amps with 12 Transistors inside are totally useless for Real Hifi, read on for why.
Why makers use ICs in Audio Stages ↑
There are many other reasons why Hifi sounds bad, cost cutting covers a lot of it as does overdesigning in search of BS high specs that appear to sound worse the more bad circuitry is introduced to "overcome" the crappiness of the design. It's all Market Forces, buyer wants "top specs", has little idea what "good sound" is & wants to pay small money. You get what you ask for, makers aren't going to make items that don't sell or are better spec than the selling price & taking chances like years ago rarely happens, which is sad. Again the lifeless thin sound that is the "tone" of the majority of Hifi since 1980 is what buyers know & why buyers of good vintage amps are so delighted hearing music with a much more natural sound.