Sound Quality Vs. Measurements

Status
Not open for further replies.
Wow!

Shock absorbers? Oil or gas? Regular or shortened and hardened?

How about the rubber? Tubeless or not? Slicks or not?

Will we need a summer and a winter thread? Hey, this sounds like a great after market sales idea - chnage your damping for just $999.99.

Low sprung chassis, adaptable so people can adjust for their hearing.

What? This is NOT the car thread? Sorry, I got my bearings wrong.

I mean look folks, I agree vibration can be a major problem, especially with high to very high gain devices such as TTs and microphones, and its elimination can be beneficial to practically all equipment. However, my experience shows that the benefits with MOST equipment are low.

Upon installing the SoundCare (from Norway) anti resonant feet on most of my equipment, I could not hear any difference, except in case of the CD player (Yamaha CDX 993), where this yielded a very slight but still audible improvement in the bass range, which became a little better defined, and more firm. The main beneficiary was the TT, of course, and I use MM; I expect that with MC, this benefit would have been even greater. I don't use mics, so I can't comment on that.

Knowing that everything vibrates anyway, I did take some precautions - the gear is on a hard, massive surface, with very good coupling to the floor. The TT is shielded from the outside world on all except the front side, which surely can't hurt. Lastly, the sound levels present in my room rarely go from semi-quiet to thundering (on occasion I do succumb to my Dark Side), typical range being 85...90 dB SPL.

On my own gear, I always put silicon paste around any larger cap's bottom, where "larger" means 100 uF and above. Power transformers are always put on rubber rings on all points, which is usually the one point, as they are mostly of the toroidal type.

How much further should I go?
 
UA 709 and 741

dNDRP.jpg


Just though these worth a look . Reading up on 741 was something I enjoyed when I did it .


As far as I can see adding a 220 R output resistor to uA 709 was the only big deal difference ( current limiting ) . I am sure at the time it seemed different . uA 709 has a respectable performance on paper considering it's antiquity .

About 20 nV noise if memory is correct with 1 MHz GBP for 741 . A pull down resistor to give a class A output if small signals ( 4K7 ) . I would imagine if using a passive ( 75 uS ) followed by active ( 318 / 3180 uS ) a moderately good phono stage for MM could still be made with it even today if using the pull down resistor on the 75 uS . Gain of active stage 16 at 1 kHz if so ( 4 x uA 741 ) . Total gain 100 . Use polystyrene for 75 uS ( 10 nF + 7K5 ) and film for the rest . It is always good to build something like this just to be certain real progress is happening . If using an IC socket the history of op amps can be listened to . Much more fun than watching soccer . I would elect Grado as a good modern PU . I like Shure 44/7,one day I intend to re-tip one with a very high quality diamond ( Expert PU in UK ) . 9.5 mV is not to be sniffed at . Shure's like 1 M ohms loading . Then adjust 75 uS to get a good 20 kHz ( do the other EQ if you like ) . That would be the PU to me with the maximum dynamic range and far less costly than my Lyra . If you have never heard a modern 44/7 in a good pick up arm you have missed something . Better than Decca London to my ears and better for the records . Stereo separation is about 15 dB down on my Lyra . Not certain it matters too much .
 
Vibration

Some RAF guy gave a talk at the Whittlebury hi fi show about measuring effects of vibration and other mysterious things .

He used a time domain before and after encoding of the information and seemed to show big differences . Talking to Richard Black afterwards he seemed skeptical . Tim de Paravacini also . He said he had tried it years ago and was not satisfied it was reality ( sorry Tim if that's not exactly true , that's how is seemed to me ) .

The RAF guy was certain it worked as it was a tool of his trade and not hi fi .

I strongly opposed this a Nordost were sponsoring it and it seemed a new expensive diversion . Very unusually for me I said no way to this . If it is provable it will grow it's own ability to prove itself . I doesn't need me . I was silenced in the meeting as I got on too well with the RAF guy who was starting to do the maths with me . The comment was approximately " for people who don't know Nigel he has very long conversations which we don't have time for " . My conversation was to be about 1 minute . It was how did we get from A to B , show me the machine so that I can build a primitive one . As far as I can see they were just doing A to D encoding and doing statistical analysis with the original . That is assuming a lot I feel .

I should add the people I work with would benefit greatly if this could be proved to be correct . I still opposed it . I much prefer we listen and decide . If not how can one be a romantic fool in this world ? I am one if asking .
 
Last edited:
Nigel, you surprise me. You seem to have had long and meaningful conversations with 'everybody and his brother' in audio, (I have had many of the same, with most of the same people), yet your 'conclusions' seem 'scattered', virtually random, each having merits in some situation, but generally very limited.
For example, bringing up the UA709 and the UA741. If you want two examples of perhaps the worst op amps for making a phono stage, you have hit the target. But do you know why?
First, the UA709 has a CLASS C output stage. The only way to fix it would be to put a resistor (at the very least) to the -supply from the output. A jfet would be much better, to remove the 'dead zone'. However, even if you did that, the internal phase shift characteristic of that design would limit your slew rate to less than 1V/us with unity gain compensation (external). What a mess! What a headache!
AND today, for about $1, more or less, you can buy a quieter, jfet input, faster, and more linear IC, with no need for external compensation caps, and short circuit proof.
The UA741 has different problems. It was a wonderful invention, back in 1969, when I was designing video servos for Ampex Research, but in trying to use it for audio, I found it SLOW, (0.6V/us), a true 'poster-boy' for TIM distortion', and NOISY, because the first stage active load adds additional noise to the input.
What is the point of bringing out these two 'dated' IC's that probably cost MORE to purchase than something more up to date?
Now, don't get me wrong, everyone. I LOVED these devices, when they were first introduced. Did you know that the original UA709's cost about $100 each? They were a 'Godsend' to military designers making compact circuitry. We used them as early as 1966, in the hundreds, at Friden, and I was put in charge of characterizing each and every one of them for offset, distortion, etc, over temperature. I tested 100 of them, with a custom tester, the schematic provided by Fairchild. They reduced a whole card of discrete parts into a little case, about the size of a typical transistor of the day. It was almost a miracle!
When the UA741 came out, we used them by the 100's at Ampex for servo design. They were great for this application, because they were input-output overload proof, and they did NOT require added cap compensation. I would still use them today for that application, but they were LOUSY audio IC amps, and it took us years to show everyone how bad they were, because some audio companies, even big ones, adapted them to save space and money, to the detriment of audio quality.
Even the major designer behind these products, Bob Widlar, in 1974, told me that these IC's were NOT designed for audio, and that we should make our own audio designs.
Today, there are at the very least, cheap and usable alternatives to these early, primitive IC's. Enough said, at the moment.
 
@JC

Exactly. Eeverything has its time, its five minutes of glory, and is then surpassed by another device.

For example, even LF 356 or 357, with their FET inputs, were great stuff - but in 1981. Today, they look absolutely antiquated, even if they are still perfectly usable.

The TLO series is another oldie but goldie - I've seen them used literally everywhere, right or wrong.

But, today I have a stock of AD817, AD827, AD829, etc. The only oldie I still have much respect for in audio is the venerable OP37, the decompensated evil brother of OP27. It stubbornly refuses to fade away and manages to sound really good even after all these years.
 
Even the major designer behind these products, Bob Widlar, in 1974, told me that these IC's were NOT designed for audio, and that we should make our own audio designs.

It is true even today, and will be true forever, like tractor is always suboptimal to be a racing car, opamp is suboptimal for audio amplifier.

Wow!

Shock absorbers? Oil or gas? Regular or shortened and hardened?

How about the rubber? Tubeless or not? Slicks or not?


However you can use mouse pads for cars, but do not expect it to last long enough.


I believe you need as much belief to believe in the big bang as to believe in the invisible guy in the sky as the origin of everything.

I believe in what you said. :)
 
Last edited:
I hope I have not been too 'strong' on this topic, but 'reason' must prevail, even though I am often attacked as 'unreasonable' on this website.
It kind of reminds me of a parallel situation, where enthusiasts are recommended to drive two (restored) autos across Europe. Let's say that I recommend a 2-CV, or a Renault Dauphine? Let's say that I found these autos enjoyable in my experience and they have quite a heritage. Well, car enthusiasts?
 
Dvv, back in 1976, I was hired as a consultant by Gibson to help with a solid state guitar preamp. We found that the LF356 sounded really good in the application, AND the designers there had come up with a very good sonic copy of a Fender tube guitar preamp. Guess what? The 'bean counters' of the company insisted on using UA741's, instead, to save money, which ruined the design, and its reputation in the marketplace. True story.
I then used the LF356 for a variable low pass filter in the Symmetry Crossover back in 1977. It was a little expensive, but it worked well enough. Perhaps you can tell me what was 'wrong' with the LF356 and why it is not seen anymore?
 
I hope I have not been too 'strong' on this topic, but 'reason' must prevail, even though I am often attacked as 'unreasonable' on this website.
It kind of reminds me of a parallel situation, where enthusiasts are recommended to drive two (restored) autos across Europe. Let's say that I recommend a 2-CV, or a Renault Dauphine? Let's say that I found these autos enjoyable in my experience and they have quite a heritage. Well, car enthusiasts?

Renault Dauphine? Heck, JC, that's mid-60ies. :D :D :D

Citrroen 2-CV wasn't REALLY a car, it was more of a bucket with a funny little engine. The thing is, those things just didn't die, they had eternity built in.

However, the two things they shared were that both were French, and that they inevitably polarized people, you either loved them, or hated them down to their goddamn carburettors. I think I'm one of the very few people who didn't love them, just liked them.

They had a certain charm, no modern car has.
 
Dvv, back in 1976, I was hired as a consultant by Gibson to help with a solid state guitar preamp. We found that the LF356 sounded really good in the application, AND the designers there had come up with a very good sonic copy of a Fender tube guitar preamp. Guess what? The 'bean counters' of the company insisted on using UA741's, instead, to save money, which ruined the design, and its reputation in the marketplace. True story.

I then used the LF356 for a variable low pass filter in the Symmetry Crossover back in 1977. It was a little expensive, but it worked well enough. Perhaps you can tell me what was 'wrong' with the LF356 and why it is not seen anymore?

Other than falling victim to fashion and no longer being "in", and being "slow" at 15 V/uS, I can't think of anything else, John.

Personally, I am never with less than 10 in my parts depot. I find it to be very reliable and very consistent, all pluses in my book.

For some reason I cannot explain, I never liked the TLO series much, so LF 356 and 357 were my workhorses. Still are.
 
Last edited:
... It is true even today, and will be true forever, like tractor is always suboptimal to be a racing car, opamp is suboptimal for audio amplifier.
... :)

Come now, my Polish cousin, be wise, never say "never".

They may not be the ideal solution even today, but let's be fair, just look at how far op amps have come just over the last 20 years.

Today, I would challenge anyone to make a discrete circuit with the overall performance of some op amps, for example, an AD 829.

Ultimately, what does "suboptimal" mean? That you can always build a better discrete op amp overall? I don't doubt it, but then take a look at how big your circuit is, how complex, and how expensive overall (i.e. including the development time)?

Have you really proved your point if it ends up costing 20, 30 or 40 times the price of an op amp?

In some applications, perhaps you have, if money does not matter. In most applications I think not, the very reason op amps are being developed today is because they are proving the point that they ARE a viable alternative in MOST, even if not in all cases.

They do, however, still have that one big limiting factor - their supply voltage range, most still being limited to +/- 15V or thereabout. I have no doubt that sooner or later, even this will be overcome one way or another.

As stated before, I find their performance way over simply acceptable once you add a discrete current booster (2 transistors, 2 diodes, 4 resistors), because the one thing they do NOT like is dealing with higher output currents, no matter what their Data Sheets say.
 
That's what I meant.

Oh. OK.

Actually, Hungarians are good mathematicians, so I can't say I'm surprised.

I don't know if you know this, but they used to have a company called Videoton, which manufactured loudspeakers. Many Brits don't know this, but a tremendous percentage of their BBC LS3 mini monitors were in fact manufactured by them and then rebadged as required. This at the time the Iron Curtain was up and well.

They made some rather good loudspeaker models, mostly for export only, so sometimes, Hungarians are surprised to see and hear their products they had no idea even existed. Ditto for Russians.
 
Status
Not open for further replies.