This is all about signal to noise ratios, at a guess the artifacts would be 60dB down or lower in level - a piece of life critical equipment would never depend on maintaining those accuracies, to be reliable.
Static only requires contact between appropriate materials to occur - add in some vibration from mechanical processes, or the air or surface movement, from the speaker drivers working, and it can only make things worse. From my perspective, I have no desire for such "funniness" to be part of the deal - but if I ignore such things I'm not happy with the sound ...
I couldn't tolerate a conventional 'distorting' valve setup - I can hear it smearing the sound, trying to 'nicefy' it - not for me, I'm afraid ...
As one of the best ‘anomalometers’, I would suggest a broadband communications receiver with many frequency bands, switchable AFC, AGC, plus a calibrated -or at least a descent- S meter.
Second alternative is a three band AM radio (LW, MW, SW)
Surveying (and recording the results) the same room at different times of a day, at different days, different weather conditions, will reveal some mysteries.
I’ve noticed 20dBm noise variations in some of the frequency bands only due to weather change.
Then, switch on and off electric and electronic appliances around the house or close to the radio. Enjoy a malt 😀
As you’ve said over and over again Pavel, cater for RF immunity.
George
Missing the point ... equipment that must work, under all circumstances, will be designed to be extremely tolerant of enviromental variations - the dumping of a bucket of water over it while hitting with a hammer at the same time, sort of thing. Generally, one doesn't listen to music in those conditions - one has time to savour the subtleties of what's going on, and hence are able to pick little irritations, flaws in what one's hearing ...
Fas42, that is also where I look for problems. I defy anyone to show me where I have deliberately added some sort of 'sound effect' due to either frequency deviation or non-linear distortion in order to add a 'sound' to the electronics.
What I try to do is 'avoid' audio distortion of all kinds, including: non-linear, linear (DA), or frequency deviation.
That is not enough, however.
There is still the power supply that never quite seems to sound as good as a battery. Why?
There might be outside interference coming in from the power line, or though the shield that is changing the sound in some subtle way.
There may be passive components that are sensitive to vibration and have a high Q that gets activated when music plays in the room, but not on the test bench.
It can be all kinds of things, but these are what counts, once you have solved the fundamental problems.
So far, almost all of these phenomena can only be heard by our ears, (if we are so inclined) and so must be evaluated by that process.
I hope, in future, like Ed Simon has demonstrated here and in some magazine articles, that I can MEASURE even more of what I hear, but until then, my ears will work just fine.
What I try to do is 'avoid' audio distortion of all kinds, including: non-linear, linear (DA), or frequency deviation.
That is not enough, however.
There is still the power supply that never quite seems to sound as good as a battery. Why?
There might be outside interference coming in from the power line, or though the shield that is changing the sound in some subtle way.
There may be passive components that are sensitive to vibration and have a high Q that gets activated when music plays in the room, but not on the test bench.
It can be all kinds of things, but these are what counts, once you have solved the fundamental problems.
So far, almost all of these phenomena can only be heard by our ears, (if we are so inclined) and so must be evaluated by that process.
I hope, in future, like Ed Simon has demonstrated here and in some magazine articles, that I can MEASURE even more of what I hear, but until then, my ears will work just fine.
Last edited:
Frank read what you stated, they would never depend on maintaining those accuracies, to be reliable...
Yet we can measure what the ears don't hear!
Oh still waiting for the proof on cable directivity.
Yet we can measure what the ears don't hear!
Oh still waiting for the proof on cable directivity.
As the Parasound JC-3 was recently discussed, I want to point out what I added to a very standard circuit (first done by Walt Jung) to make it better.
First, was power supply buffering. I added just about everything that I would into the power supply, except for size. This is what it takes to make it sound as close to a 'battery' as possible, or super quiet and floating, without adding some sort of ground loop.
The same parts that I use for EQ in the original Vendetta, and Constellation, that is, RT polystyrene caps, are used in the JC-3. I'm sure that somewhere there are equal quality good sounding caps, but I don't have the time or patience to find them. Rel RT's work, so I put them everywhere. Do they 'add' distortion? Hardly, at least I haven't measured any. Are they accurate? I try to use 1% for EQ, and that should be close enough. However, I was slightly embarrassed by the 'Stereophile' review of the JC-3 (the one with measurements) that one channel significantly deviated from the design center, which was the other channel. Perhaps the builders did not follow the spec., I will have to find out, now that I am thinking about it.
My biggest problems with this design was WHICH IC op amp to use, especially at the input?
Second, was that the builders used a lot of steel, even though the internal shields were made of aluminum. Was this a problem? Yes it was. I could hear it, so my tech and I separated the electronics from the 'steel parts' and found that it sounded acceptable. I then directed for an engineering change to aluminum for the back panel that was closely attached to the electronics. Yes, I am sort of a 'policeman' who first tries to measure, I could measure to -120dB, until now, which should be enough, and then listens, hopefully comparing it to a listening standard, like the CTC Blowtorch with Vendetta phono, and that is what I did. Finally, with much 'fussing' with different IC's and replacing the back-plate with aluminum, it was acceptable.
No, not perfect, but easily listenable. By the way, I do not get paid to put in the extra effort, I could have let the unit pass with just measurements, but I know better these days. Measurements alone will NOT insure that it sounds OK. Experience has taught me that, and that is what makes the difference, you know, the 'unmeasurable' stuff. So fas42, keep up the good work. '-)
First, was power supply buffering. I added just about everything that I would into the power supply, except for size. This is what it takes to make it sound as close to a 'battery' as possible, or super quiet and floating, without adding some sort of ground loop.
The same parts that I use for EQ in the original Vendetta, and Constellation, that is, RT polystyrene caps, are used in the JC-3. I'm sure that somewhere there are equal quality good sounding caps, but I don't have the time or patience to find them. Rel RT's work, so I put them everywhere. Do they 'add' distortion? Hardly, at least I haven't measured any. Are they accurate? I try to use 1% for EQ, and that should be close enough. However, I was slightly embarrassed by the 'Stereophile' review of the JC-3 (the one with measurements) that one channel significantly deviated from the design center, which was the other channel. Perhaps the builders did not follow the spec., I will have to find out, now that I am thinking about it.
My biggest problems with this design was WHICH IC op amp to use, especially at the input?
Second, was that the builders used a lot of steel, even though the internal shields were made of aluminum. Was this a problem? Yes it was. I could hear it, so my tech and I separated the electronics from the 'steel parts' and found that it sounded acceptable. I then directed for an engineering change to aluminum for the back panel that was closely attached to the electronics. Yes, I am sort of a 'policeman' who first tries to measure, I could measure to -120dB, until now, which should be enough, and then listens, hopefully comparing it to a listening standard, like the CTC Blowtorch with Vendetta phono, and that is what I did. Finally, with much 'fussing' with different IC's and replacing the back-plate with aluminum, it was acceptable.
No, not perfect, but easily listenable. By the way, I do not get paid to put in the extra effort, I could have let the unit pass with just measurements, but I know better these days. Measurements alone will NOT insure that it sounds OK. Experience has taught me that, and that is what makes the difference, you know, the 'unmeasurable' stuff. So fas42, keep up the good work. '-)
Not quite sure what you mean by the first sentence ...
This is a bit like the standard line about computers: they're fantastically good at doing some things which would drive us crazy, our brains haven't got a hope of competing; yet in other areas a three year old child absolutely trounces the most powerful computer in the ability to comprehend and deal with some aspect of its environment ...
Sorry Frank, you do not know what you speak about.
It is close to impossible to have any balanced technical discussion here.
Ohhh great one, is it too much to ask for pics, JC and others have posted their work, PMA amp for eg, i would love to see the inards of such nuetral electronics, quite possibly the first ...
Pics please .....
+1
It always worry's me when I sea the term "voiced" or "voicing" used in relationship to audio electronics, voicing is what Luther's and such like do to instruments not audio reproduction equipment.
Yeah , nothing like neutral electronics, if only i could see one .... 🙂
Yes, PMA, until you open up to new ideas, we will have a problem. Of course, you are an excellent design engineer, but fas42 is in the right as well. We don't want "funniness" but we have to do what is necessary to make audio equipment sound its best. That is a fundamental key to excellent design, rather than just good engineering.
Truly transparent electronics is what we design (and adjust) for. Everybody knows that, but just making it work in engineering terms does not get you there, for the most part.
A wide range shaker table might be able to excite the phenomena not seen on the test bench . Also allows you to look at chassis problem that only show up dynamically . Just a thought.
Now, what about "funniness" ? For example, I might specify a Holco resistor instead of something else for the feedback resistor from output to input in a power amp. I might make the resistor somewhat larger than necessary, as well. Works for me, every time.
Good idea, Triodehom, but I don't have a shaker table. Maybe, putting the audio device under test in the same room as a loud hi fi system might work as well.
Pics are hard to hear and no substitute to getting the posses together. Ever wonder why there are so few reports of opposing parties getting together for any disscussion about real listening to each other's standard? Some have tried, the attempts have failed. The common ground of the shows is flawed.
Sometimes this reminds me of kids in the 60's arguing over who is the best pro wrestler (male of course, the Fabulous Mulah is the ONLY female choice).
- Status
- Not open for further replies.