Not me, been there, done that. I don't get around much, anymore. Say hello to Bruno for me. Very good engineer.
john curl said:
Hi John,
These papers sound interesting. It will indeed be interesting if they are able to give relevant measurable results that expose the capacitor nonlinearity and effects of vibration. If all they do is say that a particular dielectric sounds better, or is better because it has some physical property like low DA, that will be disappointing. On the other hand, if they show some kind of measurable phenomena that demonstrates a nonlinearity or other kind of signal anomoly, that will be very interesting.
Cheers,
Bob
Bob Cordell said:
If Dr. Lipshitz let in a paper based solely on listening tests, I would be surprised. Also people forget the Pease model for DA is a linear effect causing subtle frequency response anomolies and can be made up out of a ladder of ideal R's and C's. Enough elements and you can get a match to 1ppm or better.
The paper apparently got by because it was mechanical resonance. They did imply a difficulty with ABX testing. DUH!
I find it interesting that people can't see a problem with mechanical resonances. Given that there are electric currents in the chassis--ground currents, eddy currents from stray magnetic fields from the power transformer, RF, you name it--the remarkable thing would be if there weren't problems.
Grey
Grey
I guess I don't need an AES paper about it- it's pretty easy to observe and hear the effects of vibration and noise impulses on caps. Especially the "audiophile hand-made" jobs. Make an RC series circuit, put DC across it, monitor the junction of the R and C, then tap the cap with a finger or pencil.
If I were cynical, I might attribute some reports of greater depth and warmth from designer caps to the caps having inferior microphonic properties.
Personally, I want my components to come off an immaculately-engineered assembly line.
If I were cynical, I might attribute some reports of greater depth and warmth from designer caps to the caps having inferior microphonic properties.
Personally, I want my components to come off an immaculately-engineered assembly line.
So, everyone steppin into floating Goldmund subchassis ?
(Rollin a drunk in an alley is easier if you join them and shop 'm later on)
(Rollin a drunk in an alley is easier if you join them and shop 'm later on)
scott wurcer said:
Good point, Scott. However, if that were all there was to it, we would be able to measure a frequency response anomoly. I would guess that such an anomoly due to DA in any reasonably designed circuit with any reasonable film capacitor would be very difficult to measure, much less be comparable to other frequency response anomolies in the system.
Cheers,
Bob
GRollins said:
Grey, there may be some who are disbelieving of the influence of mechanical resonances, but the cause-effect relationship is plausible. I've taken hard objects to capacitors at times. Those of us who would really like to see a measurement that illuminates the cause-effect are not necessarily dis-believers. Any time we can measure something, we learn, and learn how to make things better in a more predictable way. Just because we cannot measure it does not mean that the phenomena is not there.
Cheers,
Bob
GRollins said:
Easy test for this. Turn the volume up on your system and take a spring loaded center punch and start putting little divots in and around your chassis while you listen to the output on the speakers...
Any resonances will be clearly audible. If that's not high end enough there are companies that will sell you a calibrated version of the General Tool model. 10X for the high-end version.
All of this is based in reality, but the magnitude of the problem is the true question.
This is your cue to yell now, John.
I see a number of posts assuming I meant microphonic problems with caps.
Not so.
Take a piece of metal chassis. Create an electrical current in it via any of the methods I mentioned above. Place it near a conductor, passive device, or active device. Vibrate the piece of metal.
1) Electrical currents create magnetic fields.
2) With the vibration of the piece of metal, the magnetic field will move also.
3) What's the definition of a generator? A magnetic field and a conductor in relative motion.
Congratulations, you have just generated a pervasive low level signal in your circuit even if all the components are in perfect condition. No microphonics necessary.
This isn't murky, hand-waving, mystical stuff--this is Basic Electronics 101. A very readily understood process.
Note that this mechanism would also account for the sonic differences reported when people use non-metallic chassis materials such as wood, plastics, glass, concrete, stone, or whatever. (Yes, John, heat...I know...but I'm talking about something else at the moment.) There've been persistent reports that circuits "sound better" with non-metallic enclosures over the years. Needless to say, those who've made the claims have been mocked and laughed at (...now there's a big Duh! for you...) but it's not really all that difficult to come up with a couple of plausible ideas that don't involve the phase of the moon or being feeble-minded. I happen to favor the one above.
For the record, I'd like to state that I have not personally taken two identical circuits, placed one in a metal enclosure, the other in another kind of enclosure and listened to both for differences. With that in mind, and according to my long-standing policy, all I can say is that I have no opinion as to relative sound qualities of different enclosure materials. That doesn't stop me from thinking about it, though. (Nor does it mean that my mind is so closed that I can't think of plausible mechanisms as to why it might matter.)
And, no, my propensity for putting circuits into wood or Masonite or whatever I can get my hands on isn't related to this idea--win, lose, or draw, they're still the only materials I have on hand, regardless of philosophy.
(Man, I'm going to be really, really peeved if I finally do drop a lot of money and get my hands on a pretty chassis and the confusticated thing sounds worse than it did in the ugly form.)
Note also that the thicker the metal, the less it will vibrate. The less it vibrates, the less the magnetic field(s) will move over the circuitry inside. Again, a reasonable, logical mechanism for why high end "heavy metal" enclosures might actually sound better than mid-fi thin sheet steel enclosures.
Someone back down the line may or may not have decided that a thicker, heavier piece of metal "looked as though it was worth the money," but it may have worked to the sonic benefit of the piece anyway. Evolution didn't set out to "design" a flying creature, either. Current thinking is that wings were originally for heat radiation, then creatures discovered they could glide, and with that the whole flight process took on its own set of priorities. No, I'm not saying that amplifiers will eventually be able to fly once heatsinks evolve further--only that accidental benefits have cropped up before. Serendipity in a metal case. Who'd a thunk it?
Grey
Not so.
Take a piece of metal chassis. Create an electrical current in it via any of the methods I mentioned above. Place it near a conductor, passive device, or active device. Vibrate the piece of metal.
1) Electrical currents create magnetic fields.
2) With the vibration of the piece of metal, the magnetic field will move also.
3) What's the definition of a generator? A magnetic field and a conductor in relative motion.
Congratulations, you have just generated a pervasive low level signal in your circuit even if all the components are in perfect condition. No microphonics necessary.
This isn't murky, hand-waving, mystical stuff--this is Basic Electronics 101. A very readily understood process.
Note that this mechanism would also account for the sonic differences reported when people use non-metallic chassis materials such as wood, plastics, glass, concrete, stone, or whatever. (Yes, John, heat...I know...but I'm talking about something else at the moment.) There've been persistent reports that circuits "sound better" with non-metallic enclosures over the years. Needless to say, those who've made the claims have been mocked and laughed at (...now there's a big Duh! for you...) but it's not really all that difficult to come up with a couple of plausible ideas that don't involve the phase of the moon or being feeble-minded. I happen to favor the one above.
For the record, I'd like to state that I have not personally taken two identical circuits, placed one in a metal enclosure, the other in another kind of enclosure and listened to both for differences. With that in mind, and according to my long-standing policy, all I can say is that I have no opinion as to relative sound qualities of different enclosure materials. That doesn't stop me from thinking about it, though. (Nor does it mean that my mind is so closed that I can't think of plausible mechanisms as to why it might matter.)
And, no, my propensity for putting circuits into wood or Masonite or whatever I can get my hands on isn't related to this idea--win, lose, or draw, they're still the only materials I have on hand, regardless of philosophy.
(Man, I'm going to be really, really peeved if I finally do drop a lot of money and get my hands on a pretty chassis and the confusticated thing sounds worse than it did in the ugly form.)
Note also that the thicker the metal, the less it will vibrate. The less it vibrates, the less the magnetic field(s) will move over the circuitry inside. Again, a reasonable, logical mechanism for why high end "heavy metal" enclosures might actually sound better than mid-fi thin sheet steel enclosures.
Someone back down the line may or may not have decided that a thicker, heavier piece of metal "looked as though it was worth the money," but it may have worked to the sonic benefit of the piece anyway. Evolution didn't set out to "design" a flying creature, either. Current thinking is that wings were originally for heat radiation, then creatures discovered they could glide, and with that the whole flight process took on its own set of priorities. No, I'm not saying that amplifiers will eventually be able to fly once heatsinks evolve further--only that accidental benefits have cropped up before. Serendipity in a metal case. Who'd a thunk it?
Grey
Bob Cordell said:
There's a lot of work available on non-linear dielectrics, mostly because of their application for harmonic generation in non-linear optics. It's actually fairly troublesome to find dielectrics that are suitably non-linear at optical intensities that can be maintained for more than a picosecond at a time. (E field is around 10^8 V/M in these applications.)
GRollins said:
I wonder where the vibration comes from.
Don't tell him.Non-microphonic generation of voltage from music induced vibration. I think I'm starting to understand. How could I be so daft?
fizzard said:
I used to load some of those hi-end amps out there with a dummy load and listen to the vibrations of the heatsinks - sometimes you can even identify the song 😉
Very instructive. Yes, amps DO sound differently. At least their heatsinks do.
I'm going to try to make that presentation John mentioned.
Jan Didden
- Status
- Not open for further replies.