Hi,
Nope, not the faintest.
Remember, these are only values for ex stock. They will make any value you want (at a MOQ and price of course) all the way to 0.005% tolerance...
Ciao T
Nope, not the faintest.
Remember, these are only values for ex stock. They will make any value you want (at a MOQ and price of course) all the way to 0.005% tolerance...
Ciao T
Scott,
That is a weighted number.
What is often forgotten is the concept of critical bands in hearing theory. That allows some 30 db of discrimination in different bands.
Then allow for music content. Now normal music tends to follow the 3 db per octave rolloff above 300 hz. So looking at the ninth we would expect another 10 db. However some music has a bass boost so there can be as much as 25 db difference! (Maybe that is why some note the issues on selected musical recordings.)
When I get a chance I will post a circuit that has been proffered as an advanced audio design, that actually maximizes the feedback loop errors! (It was from one of those folks who like the sound of carbon comp resistors.)
Hint the circuit is very similar to the noise gain test circuit used to measure the performance of today's really good opamps.
ES
They look like strain gauge values to me, but then that is only speculation.
Well everyone, no matter how you look at it, many designers have found that different resistor brands have different performance characteristics, sometimes contrary to their spec. sheets. For some reason, beyond psychological, I personally believe, some resistor brands work better for audio than some others. There has been no fixed pattern as to why one brand is better than another, but it is usually consistent with the brand and the method of manufacture of that particular company.
For example, the old Holco resistors, (more than 5 years old), were completely non-magnetic (apparently gold plated rather than nickel plated end caps) and sounded 'SOFT' compared to many other resistor types. Later Holco resistors failed to sound as good, because they changed the process.
Roderstein Resistas made up to 10 years ago, or so, before they were purchased by Vishay, were an excellent combination that sounded about as good as anything out there.
They were 1%, inexpensive, (.05-.1) dollars each, had copper leads, slightly magnetic end caps, etc.
The CTC Blowtorch used them almost exclusively, with great success, ignoring even more expensive brands. Unfortunately, they have been absorbed into the Vishay machine and are no more, (exactly). Now we are trying Dale CMF resistors along with Vishay Bulk metal, for critical positions. They seem OK, so far, but we have not ended our search for the best resistor brand and type. We cannot afford to be complacent.
For example, the old Holco resistors, (more than 5 years old), were completely non-magnetic (apparently gold plated rather than nickel plated end caps) and sounded 'SOFT' compared to many other resistor types. Later Holco resistors failed to sound as good, because they changed the process.
Roderstein Resistas made up to 10 years ago, or so, before they were purchased by Vishay, were an excellent combination that sounded about as good as anything out there.
They were 1%, inexpensive, (.05-.1) dollars each, had copper leads, slightly magnetic end caps, etc.
The CTC Blowtorch used them almost exclusively, with great success, ignoring even more expensive brands. Unfortunately, they have been absorbed into the Vishay machine and are no more, (exactly). Now we are trying Dale CMF resistors along with Vishay Bulk metal, for critical positions. They seem OK, so far, but we have not ended our search for the best resistor brand and type. We cannot afford to be complacent.
I've found also that resistor in feedback sees full output voltage swing, and amp's amplification factor is directly proportional to it's value. Go figure!
Quantum plasmas extend the area of application to nano-scales, where quantum-mechanical effects gain significance. This is the case when, in comparison to normal plasmas, the plasma density is very high and the temperature is low. Then the newly discovered potential occurs, which is caused by collective interaction processes of degenerate electrons with the quantum plasma. Such plasmas can be found, for example, in cores of stars with a dwindling nuclear energy supply (white dwarfs), or they can be produced artificially in the laboratory by means of laser irradiation. The new negative potential causes an attractive force between the ions, which then form lattices. They are compressed and the distances between them shortened, so that current can flow through them much faster.
The findings of the Bochum scientists open up the possibility of ion-crystallization on the magnitude scale of an atom. They have thus established a new direction of research that is capable of linking various disciplines of physics. Applications include micro-chips for quantum computers, semiconductors, thin metal foils or even metallic nano-structures.
When ions get closer: New physical attraction between ions in quantum plasmas
The findings of the Bochum scientists open up the possibility of ion-crystallization on the magnitude scale of an atom. They have thus established a new direction of research that is capable of linking various disciplines of physics. Applications include micro-chips for quantum computers, semiconductors, thin metal foils or even metallic nano-structures.
When ions get closer: New physical attraction between ions in quantum plasmas
This resistor discussion reminds me of when a Harman purchasing agent told me that I should stop designing with 0805 resistors, as she had been assured by a salesman that they were rapidly becoming obsolete and they would be impossible to obtain. I asked her how smaller parts could be used in locations where they would run in excess of their power dissipation ratings, and she seemed puzzled. I think she gave me something like a "this one goes to eleven" look, like Nigel Tufnel in This Is Spinal Tap.
Oh well, she did drive hard bargains and saved the company money.
Oh well, she did drive hard bargains and saved the company money.
Well since things are too quiet... One of J.C.'s issues is the apparent phasing out of FET's. Why is it that FET's are still in production as parts of IC's or MOSFET's by major manufacturers but not as individual components. (Yes Scott volume is the often quoted reason, but look at some folks like That who make small quantitiies of very specialized parts.)
There are process compatibility and die size issues as well as volume, I doubt Toshiba could easily make N and P-FET's on the same die let alone npn's and pnp's.
The applications are very specialized, like IR imaging, sonar, and audio but none of them gets over at best thousands at a time.
You do have Linear Systems at least.
Think of it this way, each transient is associated with a dual skin effect transition moment. This, due to signal changing directions and transiting the DC moment.
In the resistive substrate model, you have a diffuse character and thus 1/f noise is likely the more dominant coloration. Whereas with a wire wound the material interactions likely become dominant in the expression of the transient in the 'dual skin effect transition' of a given transient. Then the complex LCR in all domains needs be looked at.
The thermal and electron noise transfer considerations needs to be analysed as a set. This is a complex and onerous task, even though the basic parameters of the interactions are obvious to both theory and evidence.
It is best solved by listening, but with an awareness of the complexities of the interaction in mind. Thus the noise levels of the given 'resistor' designs, can be analyzed in that way. since we listen and decode only transient information with our hearing, These parameters of material interaction become all we care about.
Discerning where the noise and slewing (read:hysteresis addition and reduction in character and expression of signal within the transient moment) aspects affect our ability to aurally, theoretically and electrically/molecularly understand what intelligence or signal is. Separating the carrier and medium from 'signal as intent', is the problem. Recognizing this, becomes fundamental when it comes to choices in carrier, transfer and manipulation in what are essentially plasma interactions with problematic lattice structures.
That we (or many) think that signal flows naturally through electrical wires is part of the problem. Complex LCR as a fact of Newtonian analysis ie, electrical theory..this illustrates this fundamental disconnect quite thoroughly. Plasma wants to be plasma, it wants nothing to do with a lattice structure, with the solid's so-called complex LCR bits. (complex LCR is fundamentally a Newtonian descriptor for quantum aspects- in the mass aggregate sense)
We force it to interact, in order to commit to modifications of said signal.
I think the problem with discrete fets is a bit like vacuum tubes. Back when I was a kid, tubes were everywhere. Every drugstore had a tube tester and spare tubes you could buy to keep your radio or tv going. Where did they all go?
By 1967, Ampex did not have a SINGLE tube in their normal inventory. I never saw one, there, even though I was using Dynas with KT88's and 12AX7's at home, and my TV was still full of tubes. Yet, in 1963, Ampex's finest audio master recorder used vacuum tubes, exclusively. What happened in that 4 year interval?
Well, from my perspective, bipolar transistors improved so much, during that interval, that tubes appeared to be inefficient and 'dangerous' compared to solid state. This 'progress' was apparently due to the 'cold war', mainframe computers, and the transition between Ge and Si parts. However, by 1966, we had complementary small signal Ge parts as well as Si.
Of course, we had to keep the existent tube based equipment running until it was phased out, so tubes were still made, but by 1970, we started having serious problems with replacement tubes. They were not always made to the same standard as earlier ones. Even with the GD, who cherished their Mac 3500's, this was a problem.
Discrete bipolar parts probably had their best period of growth between 1967 and 1970. After that, Motorola and just about everybody else, put their funding into IC's.
Jfets matured later, even though we had a few pretty good jfets in 1968, for example, complementary devices did not arrive until about 1-2 years later with Motorola. This is where I started with jfets. Some N ch military grade fets were also available, such as the 2N6550, at this time as well. When done right, these early fets did almost everything that today's fets do, but they WERE expensive. A couple of years later, Siliconix (Ed Oxner) came on the scene, big time, and I switched to jfets as much as I could. Siliconix was the first company, in my experience, to offer a variety of complementary jfets that were useful for audio, and some were just as quiet as the military devices.
This short 'history' does not really tell us where the majority of jfets went. Probably the military, at the time. It took a long time to get jfets into audio designs. New topologies had to be invented, exceptional parts found from the majority of parts offered, etc.
By 1978, however, the Japanese began to come out with virtually every kind and type of jfet, and they virtually took over the audio market. I suspect that most of these parts were designed specifically for audio, and now since commercial audio construction has moved away from Japan, and is usually integrated, jfet manufacturers, like Toshiba have giving up even making the devices. This is why designers and manufacturers bought up 10's of thousands of these parts, when they were still cheap to buy, and continue to make circuits with jfets. Mosfets are another story.
By 1967, Ampex did not have a SINGLE tube in their normal inventory. I never saw one, there, even though I was using Dynas with KT88's and 12AX7's at home, and my TV was still full of tubes. Yet, in 1963, Ampex's finest audio master recorder used vacuum tubes, exclusively. What happened in that 4 year interval?
Well, from my perspective, bipolar transistors improved so much, during that interval, that tubes appeared to be inefficient and 'dangerous' compared to solid state. This 'progress' was apparently due to the 'cold war', mainframe computers, and the transition between Ge and Si parts. However, by 1966, we had complementary small signal Ge parts as well as Si.
Of course, we had to keep the existent tube based equipment running until it was phased out, so tubes were still made, but by 1970, we started having serious problems with replacement tubes. They were not always made to the same standard as earlier ones. Even with the GD, who cherished their Mac 3500's, this was a problem.
Discrete bipolar parts probably had their best period of growth between 1967 and 1970. After that, Motorola and just about everybody else, put their funding into IC's.
Jfets matured later, even though we had a few pretty good jfets in 1968, for example, complementary devices did not arrive until about 1-2 years later with Motorola. This is where I started with jfets. Some N ch military grade fets were also available, such as the 2N6550, at this time as well. When done right, these early fets did almost everything that today's fets do, but they WERE expensive. A couple of years later, Siliconix (Ed Oxner) came on the scene, big time, and I switched to jfets as much as I could. Siliconix was the first company, in my experience, to offer a variety of complementary jfets that were useful for audio, and some were just as quiet as the military devices.
This short 'history' does not really tell us where the majority of jfets went. Probably the military, at the time. It took a long time to get jfets into audio designs. New topologies had to be invented, exceptional parts found from the majority of parts offered, etc.
By 1978, however, the Japanese began to come out with virtually every kind and type of jfet, and they virtually took over the audio market. I suspect that most of these parts were designed specifically for audio, and now since commercial audio construction has moved away from Japan, and is usually integrated, jfet manufacturers, like Toshiba have giving up even making the devices. This is why designers and manufacturers bought up 10's of thousands of these parts, when they were still cheap to buy, and continue to make circuits with jfets. Mosfets are another story.
- Status
- Not open for further replies.