Sound Quality Vs. Measurements

Status
Not open for further replies.
Asked why NE5534 and 2N3055 ? Apologies to John . You are absolutely right in the strictest sense . I ran a course in mechanical engineering . It always seemed to me the best engineers could work with the most basic stuff . We did get very good marks . I think the reason is all of the projects I used were for my real world work . Much of it was vibrational analysis . The guys had to learn electronics to use strain gauges etc . I never helped with that as they didn't need help . They found jumping disciplines very easy .

Now a slight leap of faith . Within reason it should be possible to upgrade any good design . For example no high performance engine I have seen in common use is distinctly different from the Zeppelin engines of 1920 . Already 4 valves per cylinder and aluminium . The engines which preceded it were primitive . Aero engines took years to get further as the Zeppelin had one advantage . Engine failure would not cause it to immediately crash . In air repairs were even done ( plugs , magneto , oil leak , carb failure ) . The big deal was metallurgy , the Russians are the absolute masters of that ( railways in Siberia , 3 times the exit temperatures on rocket engines ) . The Zeppelin had no choice , it had to be lightweight . One could say it was Germanium in what it used .

The funny part of this story is many motorcycles had 4 valves in the 1920's ( aero inspired ) . They also had iron pistons . I won't work . Rudge were about the first to make it work . 5534/ 3055 are not iron pistons .

If you like the device is the metallurgy ( op amps and transistors ) . Topology is function . Just like for a chef the best ingredients work better . However a good chef can make supermarket pasta taste great . Maybe he adds fresh truffles ( BYV28 diodes ) ? I can make the best ingredients taste at best OK . I hope the point is made . It is to test the designer .

My observation is my students got wrapped up in what they did . Being cheap to do they can risk having a go with my competition ( 5534 less than 50 cents often 3055 less than $1 ) . All people are curious by nature . the evolution of a design is inevitable . I know of no one who who would stop where the competition rules put them . The thing is we might get some of our children away from computers and using a soldering iron . If using valves PCL 86 and an op amp would be where would start . The Spud amp completion is a bit daft . I think as I have said before PCL 86 is not allowed . Sad as I was onto a free holiday . I have knocked together a Spud amp . It has two valves . It will have 3 as I see something wonderful emerging . To be honest the one thing I learnt form it is good amps are good amps . I was hoping it would improve MP3 material . It didn't . It has all the below 1 watt measurements one could hope for . Now 1 to 8 watts must improve . My quest is to meet DIN 45500 whilst not spoiling how it sounds up to 1 watt . I suspect MP3 is about DAC's . I have a Sony Mini disc which does work , it can be done . That's another point . I use a mildly modified Quad 33 303 for my TV . It never fatigues and is OK with MP3 . Before anyone says because it is bandwidth limited I will say nonsense . I have tried it . Generally bandwidth limiting makes it worse if the amp is suitable to go further .

The final part of the test would be a blind listening test against the latest Quad amp . Not that it is in anyway the worlds best . I would use it because they were always the ones who insisted all good amplifiers sound the same . I don't totally believe in blind tests . Many can never relax when doing it . They feel tested rather than helping to test .

Thanks for all the Wein bridge stuff . My turntable PSU uses a SVF with 3 rd harmonic null ( Rosen inspired ) . The outputs of the sections summed in a 91K and 10 K before a buffer amp . That slightly increase the 5 th harmonic , however that is already very low . My little Wein was the oscillator of the turntable in the past . Thorens used a Wein in the 1970's on TD125 . The motor is special to work at low voltage . A work of genius . LP12 should have done the same . It would be ULTRA easy to wind the coil . It could be done with a hand drill . The Airpax motor if filed will solder ( steel can be soldered , use plumbers flux and 60/40 PbSn ) . The motor is crimp riveted . I have repaired many .


For your amazement look at this ( 15 years before a DC 3 , before the Ford Tri-motor ) .
https://en.wikipedia.org/wiki/Zeppelin-Staaken_E-4/20
 
One student of Stanford University decided to build a networked computer long time ago from parts that can be bought in a store... Another student compiled Berkeley Unix for it. It become a foundation of Sun Microsystems...

And when the first student saw the third student and asked him what was he doing, the third student said: I'm drinking.

When the second student a few years later saw the third student and asked him what he was doing, the third student replied: I'm still drinking.

When the three students met up last year, the first two came in cheap, bland family station wagons. The third came in a Ferrari. The first two asked him what happend, and he replied: I sold off all those bottles.

When I was in Babson college in Boston in 1991, they showed me their Hall Of Fame corridors, with portraits of known students of theirs. True, I did see Charles Wang (Wand Computers) and Ken Olsen (Digital, designed VAX), but most I didn't recognize, even if I was subscribed to Business Week. It turned out that most of the others were in services.

THAT'S when the US economy atmosphere changed, and quickly moved away from manufacturing towards services, especially financial services.

Wave, think - if a bright student then could do all that with the parts available then, try to imagine what such a student could do with the parts available today. It boggles the mind.

But the fact is that most western economies are turning towards trading and services, and away from manufacturing. More money is used to make the investors happy with dividends than in research. Except for the military, they never seem to lack the resources for whatever they can dream up or what can be dreamt up for them.

Aready there are more investors than sustainable projects in this world, not because it's literally so, but because investors want incredible returns in almost zero time. Hence, we have this crisis, which is taking longer than any other ever - the next one will be even worse.
 
I effectively was at Oxford University for 23 years . Long after I finished studying my customers kept me up to date . One gets a feel for the buzz of invention and the ones who do it . Alas more and more Audio is a little backwater . All students had 10 % discount . 90 % of my customers were students . You can work out that we must have been the cheapest truly high end shop in the world . Some of our students could afford Naim , Quad and big Tannoy's . That was before silly hi fi ( jewelery ) . I remember Linn and their CD player made from solid billet aluminium . Sort of good / silly . Now Linn can not be bothered to make a CD player . For pride alone they should . The old one with a new chip set would do . John your right , these are easy upgrades done for us by the chip makers . That's what Linn should have done .

BTW . The company Westwood's had a retail , wholesale components and professional video side . I was supposed to be a service engineer whilst I studied . I was asked if I could help with the retail . I fell in love with it and ran it until 1997 . Student grants were cut and parking was impossible , then the rent went up . If I win the lottery I will reopen the store . It never made much money and it would make even less now . The business dated back to 1937 . It sold Sony in 1962 when there was no official importer ( was always invited to all Sony events because of that ) . When I was 4 years old I went in with my dad to buy parts . It was as near to a boys paradise as it was possible to know , an Aladdin's cave . Behind the scene's even more so . Stock which would now be worth a fortune , especailly the valves .

I met the designer of FAL Japan . Almost an identical story .
 
Last edited:
Anyone actually measure a 2N3055? The 2N3055 was a Hometaxial design. As this process is no longer used everything made today is not really designed the same. That means the transistors sold as 2N3055 now have secondary breakdown issues and higher Ft than the actual first parts.

For a while the specifications were so low by more modern standards that many manufacturers would label any parts that did not meet better spec.s as 2N3055.
 
Hi Simon . I can still get 2N3055 H . I am told it is tougher , however some say the other version sounds better . I know some 3055 go to 3 MHz these days . TIP 3055 I was told did ( A&R A60 versions did ) . I guess some are other transistors inside them now ? That could be part of the inventive skills required . sourcing and testing . I would say using better to start the design and see what 3055 change when used . On reflection I think TO3 should be specified . Some might even try using better and printing new numbers ( cheats ) . An eagle eye needed . Faking a 5534 would be harder .
 
Here is what I think is the original data for a 2N3055. Note that it has a gain of 20 at 10 khz!
 

Attachments

  • untitled.JPG
    untitled.JPG
    97.2 KB · Views: 133
The only time I designed with a 2N3055 was for a motor driver for a servo driven capstan. Worked just fine.
Dyna started with the 2N3055, but I have only used the ST120 for a motor drive, never for audio. Perhaps, I am a 'snob', but what is the point?
There have been better complementary parts since 1967, from Motorola, so I started there. Why bother, I thought at the time, I already had a pair of Dyna MK3 power amps, with a matched pair of KT88 tubes, why use anything but the most elegant parts to try to make something better than what I already have? Even that took some doing.
When it comes to local prices and availability. Perhaps both of you are right.
When I lived in London in 1976, I used Motorola output devices MJ15002,3. They were not too hard to source.
For everyone else, my comments are based on 'What do you want to do with your time?' Do you want to fix Yugos, or old Austins, because they are available locally, cheap? Or do you want to work on old Porsches, or the like, because they promise better performance, for the time and trouble that you put in, fixing them up?
 
hfe of 20 is fine . 2N3055 goes down to 5 at 1 ohm . Uses a 3055 to drive a 3055 if so . Gets you 25 ( use H version ) .

Below biased at 10 mA 2N3055 using triples . The bias isn't very critical to get this result . Other tests show near perfect waterfall distortion ( as I have seen it written , exponential harmonics ) . Like a SE tube amp , only the staring point is 20 db lower or better ( often 40 dB ) . Jean Hiraga was saying this in the 1970's . Many still do not mention it . Hiraga says we hear a diversion from exponential as distortion . The actual level can be fairly high and seem undistorted if following this rule . If Hiraga still says it I don't know . I beleive it as it goes with my observations . Hirage stated distortion must have qualities like real music to be accepted . Thus an amp with only 0.05 % fifth harmonic sounds worse than one with 1 % linear ( exponential ) harmonics . Hiraga did say he wasn't sure if linear better than exponential . He favoured the latter .

This is the 1965 origin ( 1967 production ) amp using 2N3055 . 45 W 20 Hz to 45 kHz .
http://2.bp.blogspot.com/_yc3hhZdhO...A3k/xe7NKS741sA/s1600/Quad+303+distorsion.bmp

Going with Johns Porsche theme . My friend spent lets say $20 000 20 years ago doing up his VW .convertible ( like a certain person used in 1942 , not the camper van ) . Having driven it I was mighty impressed . He had used all Porsche parts , however power was standard . He reasoned that the horsepower often is the last thing that needs uprating . It remains in my memory as fantastic . The Porsche I went in recently could learn from it . Very fast and nothing else . Amplifiers should be upgraded from the power supply first . For me that is transformer first . No harm in better transistors to kick it off . I was heavily influenced by Julian Vereker on this . Nothing has changed my mind . Naim transformers were differently wound . Some would see them as defective ( no ) . They would certainly blow fuses and hate DC on the mains . Naim used BDY 56 , a super version of 3055 ( 10 MHz ) , very sweet sounding . Later Naim used Naim transistors ( no idea what ) .
 
Last edited:
DVV your are right about the military . When I said I needed some 2SD756 gain > 400 I was told " my friend can make those" . Knowing how genuine this friend of mine is I said how come ? It will be put in the books a a military device ! I hate to think what it would have cost . I was confidently told it would at least equal the original . Not the UK tax payer , I dare not say which country . How much should I pay ? " Oh $1.61 ( a pound ) each would be fair " .
 
The only time I designed with a 2N3055 was for a motor driver for a servo driven capstan. Worked just fine.
Dyna started with the 2N3055, but I have only used the ST120 for a motor drive, never for audio. Perhaps, I am a 'snob', but what is the point?
There have been better complementary parts since 1967, from Motorola, so I started there. Why bother, I thought at the time, I already had a pair of Dyna MK3 power amps, with a matched pair of KT88 tubes, why use anything but the most elegant parts to try to make something better than what I already have? Even that took some doing.
When it comes to local prices and availability. Perhaps both of you are right.
When I lived in London in 1976, I used Motorola output devices MJ15002,3. They were not too hard to source.
For everyone else, my comments are based on 'What do you want to do with your time?' Do you want to fix Yugos, or old Austins, because they are available locally, cheap? Or do you want to work on old Porsches, or the like, because they promise better performance, for the time and trouble that you put in, fixing them up?

Careful with the Yugo, John. :D :D :D

But seriously, you are right, the Yugo was a pile of junk. And, in all fairness, you have no idea what a pile of junk. The ones which went for exports had zinc plated bodywork, the local ones didn't ans started to rust from the factory, literally. The export models used a version of the bodywork which was stiffened up for FIA Group 4 saloon racing, the local ones were as soft as paper.

SY, I am NOT talking cars here, I am trying to point out that everything that look like something may in fact be way worse that it looks, because of the time honoured communist principle to have two series of whatever, one for the domestic market (usually the worst of junk), and one for exports (usuall much better than the dometic one, even if still junk overall).

But, as with everything, even the worst of the junk can be turned into a little pearl, it's just a matter of time, knowledge deployed and money sunk into it. True of cars, true of electronics.

I have pulled a few funny ones in my time. Probably my best was a German made Wega receiver ( http://wega.pytalhost.com/1975/wega-04.jpg ), which used - you guessed it! - 2N3055 in its output stages. Nevertheless, as long as you didn't aks for too much power, it made some reasonably good music. Changing 2N3055 for MJ211xx, as well as a few critical resistors from carbon film 5% to metal film 1%, did make a hell of a difference even without fiddling with the compensation which was done to suit 2N3055. It was a friend's receiver, so no money was involved, but the pleasure of improving was all that much greater. He still has it to this day, it never broke done once.

In that year, 1975, Sony acquired Wega lock, stock and barrel, and thereafter they repackaged Sony products, so that was the last of the real Wega.

Anyway, in my book, 2N3055 has its place in the Hall of Fame, but it is most definitely history I have no wish of revisiting.
 
DVV . Funny thing is some fit carbon composition as they find metal films sound harsh . I must say the best resistors I ever used were metal foil . Too expensive to use generally . I am still desperately seeking a better basic resistor . I still use MRS 25 . It's OK .

Tantalum film ? I never tried them . The capacitors put me off .

BTW . " Careful with that Axe Eugene " as the song says . Many amps osculate when better transistors are fitted . Upgrading a Goodmans Module 80 , as our engineer Leon Hendrik's said " Oh the one with smoke emitting diodes ( LED ) " the original driver had to be used as nothing seemed the better replacement . The one mod that did work was replacing the driver emitter diode with a resistor . It gave identical DC points . however it sounded better ( 56R ) . We did an IM distortion test and sure enough it was better ( some many dB ) . I see people diode bias valves . Not me even it it does work . The amp sounded wonderful after it's upgrade . It was a customer who insisted . My brother did it and got £160 for his trouble . All capacitors were replaced except some ceramics . Films were used were possible and not smaller than 100 V types , many polystyrene . To be honest a Yugo of an amp . Brave customer and his bravery paid off . He liked it for how it was domestic and not intrusive . He had a Garrard 401 so was worth the effort . My old Boss though the ultimate deity drives a Land Rover and a Transit van ( ? , I'm mystified also ) , a Harley Davidson and uses a Garrard 401 .I sort of hope my old Triumph Tiger is in his collection .
 
Last edited:
I am tempted to say I always was the light when power supplies . I am certainly an evangelist . I suppose like you Dvv I had to make things work and we were very poor in those days . As a proportion of our money hi fi was something we spent money on more then than now .

My local tool shop Screwfix has a HD 320 GB hard drive PVR satellited receiver and 4 way LNB and dish for $159 . In the old days it would not buy you the cheapest pick up , £7 in 1974 ( basic wage was £14 for a kid ) , My discounted Akai AA8080 was $324 of now money or about 40% the price of a BSA 650cc motorcycle new . I was 14 at the time ( 1970 ) . I went to the shop in Silverdale Road Hayes ( Comet , near EMI and where Blumlein worked , Jordan watts also ) . I went there because no one in Oxford believed I had the money . I went to the counter and it was handed to me " go over there to pay " . the assistant totally trusted me . It was to be the Armstrong 625 , I couldn't get one ( I have one now ) . That Akai was a work of art , sonically not so hot . The FM was fantastic . I can remember on the train thinking do people know I have all this money ? Also on returning will they think that I had stolen the Akai ?
 
Last edited:
DVV . Funny thing is some fit carbon composition as they find metal films sound harsh . I must say the best resistors I ever used were metal foil . Too expensive to use generally . I am still desperately seeking a better basic resistor . I still use MRS 25 . It's OK .

Tantalum film ? I never tried them . The capacitors put me off .

BTW . " Careful with that Axe Eugene " as the song says . Many amps osculate when better transistors are fitted . Upgrading a Goodmans Module 80 , as our engineer Leon Hendrik's said " Oh the one with smoke emitting diodes ( LED ) " the original driver had to be used as nothing seemed the better replacement . The one mod that did work was replacing the driver emitter diode with a resistor . It gave identical DC points . however it sounded better ( 56R ) . We did an IM distortion test and sure enough it was better ( some many dB ) . I see people diode bias valves . Not me even it it does work . The amp sounded wonderful after it's upgrade . It was a customer who insisted . My brother did it and got £160 for his trouble . All capacitors were replaced except some ceramics . Films were used were possible and not smaller than 100 V types , many polystyrene . To be honest a Yugo of an amp . Brave customer and his bravery paid off . He liked it for how it was domestic and not intrusive . He had a Garrard 401 so was worth the effort . My old Boss though the ultimate deity drives a Land Rover and a Transit van ( ? , I'm mystified also ) , a Harley Davidson and uses a Garrard 401 .I sort of hope my old Triumph Tiger is in his collection .

What's "better"?

But I know what you mean. And I am hardly surprised. I see people modding their amps, taking out 100 or 130 W devices and inserting 200W devices instead. Mostly it works, but just as often, if not more often, they say they gained on sound quality, but not on quantity. Fine, I say, have you modfied the protection circuits, to suit more powerful output devices? Hell no, I ain't messin' with that, most say.

The problem with far too many mods is that they are superficial, things have been changed, but it's mostly the placebo effect at work, unless the original was like 40 years old and you stick in some yesterday's transistors, but then you risk a bit too, the original couldn't hit 1 MHz, and the new one does like 60 MHz.

Mostly, I agree with your summary, it's the power supplies which should be modded first, as that's where you maximize your gain, your bang per buck/pound/euro. A better quality transformer, preferably with more VA, with better rectifiers and better and bigger caps will defintely make each and every amp sound better, the question is only by how much.

Sometimes, yopu do get surprises, no matter what your experience is. My greatst ever suprise was when I refreshed my old HK 6550 integrated amp, put in new capacitors all around, but stuck to the original large 15,000 uF caps by some Korean manufacturer, purchased as spaere parts from HK. It worked, of course, the amp was purchased in 1993, after all, but it was hardly stunning. Then I did what I had been wanting to do for a long time, I swapped its no name volume pot with an Alps Blue, and then it took me by surprise.

If anyone asked me whther changing one pot could make much difference, I'd say it should make some difference, but no stunner. Well, I was wrong, it CAN be a stunner. I still find it hard to believe how much better it sounds now, from a solid but unexciting it had gone up in one step, to solid and exciting, since it gained sound stage depth and height I never knew it could do. I should have done that a long time ago.
 
BTW, Nige, I was never poor, it's just that I had no sources to buy from.

In 1970, you say the basic wage was 17 pounds; that year, one trimester in my school in Somerset cost 950 pounds, or 56 times that, and that's just basic costs, pocket money, clothing money and travelling money was extra. That is hardly poor. (For the Americans here, in those days 950 pounds was around $2,300, about the price of a reasonable new car in those days).

Whether that money was squandered or not is not for me to judge.

Today, I have quite a number of vintage items basically sitting in my cupboards simply because I cannot listen to all of it at once. That also is hardly poor. And I sure am happy.

The truth is that I have always had more than most in Europe, about as much as US family well off fared in those days - not rich, but well taken care of.
 
Nah. Amplifiers have to be upgraded from set of criteria of optimization. Then you can see what in it can be changed in order to meet the requirements. Build what you want in imagination, and compare with what you have. The remainder is what you need.

Well, for PSUs that's easy - there is no such things as too much quality and good quantity.

Put in the biggest tranfrmer you can given the space, and make it as good as you can afford and/or care about the device being upgraded. Ditto for capacitors.

Wave, you simply cannot fail with that simple formula. THEN, after you have actually heard the improvements, decide whether you want to go on or not.
 
Well, for PSUs that's easy - there is no such things as too much quality and good quantity.

Put in the biggest tranfrmer you can given the space, and make it as good as you can afford and/or care about the device being upgraded. Ditto for capacitors.

Wave, you simply cannot fail with that simple formula. THEN, after you have actually heard the improvements, decide whether you want to go on or not.
Put marble feet and wooden knobs. " THEN, after you have actually heard the improvements, decide whether you want to go on or not."
I like your hypnotic formula. It is classical, contains nice presupposition about "after you have actually heard the improvements". :D
 
Status
Not open for further replies.