Umm, the linear versus non-linear discussion appears to be rather academic to me.
But consider the following: If you add a signal precisely with a delayed version of that same signal you get something that REMOVES harmonics (in stead of adding harmonics). When applied to a music signal this will sound rather distorted to us. But still it is a linear system...
But consider the following: If you add a signal precisely with a delayed version of that same signal you get something that REMOVES harmonics (in stead of adding harmonics). When applied to a music signal this will sound rather distorted to us. But still it is a linear system...
No they are not, as long as you don't try to operate them near saturation. Any component can be forced to become nonlinear if operated outside of it's normal intended operating range. Apply enough voltage to a cap to get it to arc internally and it will show nonlinear performance big time. The active parts are the big generators of nonlinear distortion in analog circuits, not the passive ones,in general.Steve Eddy said:
Hehehe. Ok, how 'bout audio transformers? Are they designed to have nonlinear performance? Or are they special cases too?
se
BZ
Pjotr said:But consider the following: If you add a signal precisely with a delayed version of that same signal you get something that REMOVES harmonics (in stead of adding harmonics). When applied to a music signal this will sound rather distorted to us. But still it is a linear system...
I must have forgotten what I learnt in the system theory
courses, since as far as I remember a system with time delay
(that is a frequency independent delay, in contrast to phase
shift) is not linear.
Pjotr said:
Right, but you missed my point a bit I think. Such a system as I described generates a comb filter. It can be build by components that are perfectly linear and even completely passive.
OK, I don't quite know what a comb filter is, although I have
heard the term before. But one can build a delay line with
only linear components, of course, so I think I see what you
mean.
Pjotr said:
Right, but you missed my point a bit I think. Such a system as I described generates a comb filter. It can be build by components that are perfectly linear and even completely passive.
OOOOHHHHH..now I get it..
Talk about ya inna godda da vita...
Couldn't figure your post out..geeze, ya shoulda said phasing/flanging....den, I wooda understood..
Cheers, John
BTW...did that for years back in the 70's, as a DJ in a nightclub..you should hear what that sounds like in the headphones, with each turntable feeding one ear..really weird how the sound source sweeps right through the head..
Hi,
While they're bot designed for non-linear performance, at some point they will go into non-linear behaviour.
Coupling xfomers, audio chokes and basically anything with a magnetic core can go into hysteresis and therefore distort the signal passing thruogh it rendering it non-linear.
FWIW, DA and DH are not the same. One being an artefact of the other.
DA is important in capacitors and ICs, the better the insulator and the wire it's insulating the more likely the signal sent through it will come up unaltered at the other end.
Still, there are other factors at work as well but it is perhaps better to stick with the more important ones for the moment.
Cheers,
While they're bot designed for non-linear performance, at some point they will go into non-linear behaviour.
Coupling xfomers, audio chokes and basically anything with a magnetic core can go into hysteresis and therefore distort the signal passing thruogh it rendering it non-linear.
FWIW, DA and DH are not the same. One being an artefact of the other.
DA is important in capacitors and ICs, the better the insulator and the wire it's insulating the more likely the signal sent through it will come up unaltered at the other end.
Still, there are other factors at work as well but it is perhaps better to stick with the more important ones for the moment.
Cheers,
Re: transformers & distortion
Hi,
It is not my intention to be arrogant but it cannot do much harm to think a bit further when people state that trannies are linear parts.
The source driving the trannie needs also to supply the magnetization current of he core. Since this magnetization is far from linear (also, or even especially, at low magnetization for some OT’s) the current needed is also far from linear. When you drive a trannie with current this non-linearity shows up more exaggerated in the audio signal. Something to watch if you have a pure penthode PP output stage for instance.
Cheers
Claude Abraham said:
Hi,
It is not my intention to be arrogant but it cannot do much harm to think a bit further when people state that trannies are linear parts.
The source driving the trannie needs also to supply the magnetization current of he core. Since this magnetization is far from linear (also, or even especially, at low magnetization for some OT’s) the current needed is also far from linear. When you drive a trannie with current this non-linearity shows up more exaggerated in the audio signal. Something to watch if you have a pure penthode PP output stage for instance.
Cheers
fdegrove said:
Quoting from that reference we learn that
However, this is as far as I can see exactly one of the several
physical phenomenae contributing to dielectric absorption
as explained by Kundert (http://www.designers-guide.com/Modeling/da.pdf). I do admit, however, that I
did not take the time to reread Kunderts paper to check
that my memory of what it says is correct.
re: re: transformers & distortion
That's what I thought you meant. You are, of course correct regarding the non-linearity between the V vs. I of a transformer. For any ferromagnetic core material, the B-H, and hence the V-I relationships are indeed non-linear. However, when a transformer is described as "linear", it is usually understood that the input-output relationship is that of voltage vs. voltage (common voltage transformer), or current vs. current (current transformer). As long as the operating flux density is kept below saturation, it is indeed truly linear. A transformer is commonly used to scale voltages or currents. It is not used, to the best of my knowledge, as a V to I (transadmittance) or I to V (transimpedance) element. Attempting to do so would indeed result in large distortion due to the V-I and/or I/V non-linearity. Best regards.
Pjotr said:
That's what I thought you meant. You are, of course correct regarding the non-linearity between the V vs. I of a transformer. For any ferromagnetic core material, the B-H, and hence the V-I relationships are indeed non-linear. However, when a transformer is described as "linear", it is usually understood that the input-output relationship is that of voltage vs. voltage (common voltage transformer), or current vs. current (current transformer). As long as the operating flux density is kept below saturation, it is indeed truly linear. A transformer is commonly used to scale voltages or currents. It is not used, to the best of my knowledge, as a V to I (transadmittance) or I to V (transimpedance) element. Attempting to do so would indeed result in large distortion due to the V-I and/or I/V non-linearity. Best regards.
Hi Claude,
To some extend that is true. But in practice trannies are driven neither by a voltage source nor are they driven by a current source. OT’s are driven by a source with substantial Ri. Also they are not loaded secondary open circuit or short circuit, but they drive a load impedance which is often complex. Then the nom-linear magnetisation current has its influence, especially at lower frequencies.
And yes a penthode output stage is more like a I/V converter than a V/V converter.
Cheers
To some extend that is true. But in practice trannies are driven neither by a voltage source nor are they driven by a current source. OT’s are driven by a source with substantial Ri. Also they are not loaded secondary open circuit or short circuit, but they drive a load impedance which is often complex. Then the nom-linear magnetisation current has its influence, especially at lower frequencies.
And yes a penthode output stage is more like a I/V converter than a V/V converter.
Cheers
fdegrove said:
Ok. I read it, but I don't see how this explains that DA, DH and soakage are different things.
They mention leakage current, but I don't see how leakage current can be considered the same as either DA, DH or soakage.
With leakage curent, you effectively have the same amount of charge entering the dielectric as leaving it so that at any given time the net charge of the dielectric is zero.
Whereas DA, DH and soakage have to do with what amounts to a residual polarization of the dielectric, much akin to the residual magnetization of ferromagnetic materials which causes hysteresis in them.
se
Hi,
In a way, yes. Saying they're all the same is not correct though.
But we're still miles away from diodic effects in cables wouldn't you agree?
So, where could this notion of diodic effects come from?
As I think you know the answer, I wonder why you ask.
Isn't it ironic how most cable manufacturers have tried to avoid the effect by striving to make high purity wire and now, so many years after the event, you bring it back up again?
Some posts back I mentioned the effect of the wire/insulator interface, guess that slipped the attention of the crowd too?
The term I used, I think, was interdielectric barrier? Not sure.
Anyway, this theory doesn't apply to magnet wire anyway so somehow, maybe unconsciously, you're on the right track anyways.
Ever gave it an indepth thought on as to why magnet wire sounds so great despite inferior dielectric specs?
Cheers,
In a way, yes. Saying they're all the same is not correct though.
But we're still miles away from diodic effects in cables wouldn't you agree?
So, where could this notion of diodic effects come from?
As I think you know the answer, I wonder why you ask.
Isn't it ironic how most cable manufacturers have tried to avoid the effect by striving to make high purity wire and now, so many years after the event, you bring it back up again?
Some posts back I mentioned the effect of the wire/insulator interface, guess that slipped the attention of the crowd too?
The term I used, I think, was interdielectric barrier? Not sure.
Anyway, this theory doesn't apply to magnet wire anyway so somehow, maybe unconsciously, you're on the right track anyways.
Ever gave it an indepth thought on as to why magnet wire sounds so great despite inferior dielectric specs?
Cheers,
fdegrove said:
Ok, if it's not correct, then explain what the specific differences are. Otherwise, I've seen no evidence that those three terms aren't all describing the same phenomenon.
Sure.
Beats me. People in this industry make stuff up out of thin air all the time so the notion could have simply come from thin air.
I ask because all I've seen so far are empty claims by high-end audio cable manufacturers.
I don't know if they're trying to avoid the effect or simply trying to gain some marketing advantage. Again, people in this business make stuff up out of thin air on a regular basis. And I bring it up still because the claims are just as empty today as they were back then. You'd think by now they'd have more than a bunch of empty claims made in their marketing literature.
All you did was mention it. You didn't seem at all willing to discuss it from any sort of technical standpoint.
Dunno. Suffice to say that "interdielectric barrier" doesn't get a single hit on Google.
Why wouldn't it apply to magnet wire?
Sure. But since included among any possible reasons are psychological factors, getting any meaningful answers would first involve ruling out psychological factors by means other than handwaving, empty claims, and sheer ego.
I've discussed how magnet wire can be technically superior in spite of inferior dielectric specs, but that's just a purely technical issue and a far far cry from whether it has any actual audible effect.
I just go with what sounds best to me and refrain from making any sort of objective claims as to why.
se
Hi,
Well, I can't remember where I read that but I had the impression that it was just an excuse/explanation to promote the use of high purity metals in audio wires.
Naturally microdiodes do exist but how they could possibly have any effect in a cable of decent quality is simply beyond me.
If we know at what infinitissimally small levels these operate and knowing that they'd be randomly aligned along a conducting wire that actually would short them all out, I wonder how they could possibly affect an AC signal at AF.
While it's nothing really new it's not often been implemented in high end audio cables by the big players.
In fact I recall bringing it up last year, in a slightly different context, and recall that you argued it just couldn't happen.
Let me just say it's a way of neutralising static electricity caused by current through the wire/insulator interface.
The same technique is used by manufacturers of high pulse capacitors and has been known in tube manufacturing since the Thirties.
Because it's an airtight insulator and adding an intermediate layer would break that.
Also the enamel coating doesn't build up any charge to the best of my knowledge.
Good on you.
I just hope to have lifted the veil a little...
If you'd like to read up on microdiodes or single electron transfer mechanisms you could find a few references on the web on that for sure.
While it makes for great bedtime reading on a sleepless night I very much doubt the effect is audible to even the most golden eared amongst us but what do I know anyway....??
As is often the case when advancing the state of the art it's when you add up all attention paid to even the most seemingly trivial details that you can end up with an advance over the commonly known possible.
Cheers,
Well, I can't remember where I read that but I had the impression that it was just an excuse/explanation to promote the use of high purity metals in audio wires.
Naturally microdiodes do exist but how they could possibly have any effect in a cable of decent quality is simply beyond me.
If we know at what infinitissimally small levels these operate and knowing that they'd be randomly aligned along a conducting wire that actually would short them all out, I wonder how they could possibly affect an AC signal at AF.
While it's nothing really new it's not often been implemented in high end audio cables by the big players.
In fact I recall bringing it up last year, in a slightly different context, and recall that you argued it just couldn't happen.
Let me just say it's a way of neutralising static electricity caused by current through the wire/insulator interface.
The same technique is used by manufacturers of high pulse capacitors and has been known in tube manufacturing since the Thirties.
Because it's an airtight insulator and adding an intermediate layer would break that.
Also the enamel coating doesn't build up any charge to the best of my knowledge.
Good on you.
I just hope to have lifted the veil a little...
If you'd like to read up on microdiodes or single electron transfer mechanisms you could find a few references on the web on that for sure.
While it makes for great bedtime reading on a sleepless night I very much doubt the effect is audible to even the most golden eared amongst us but what do I know anyway....??
As is often the case when advancing the state of the art it's when you add up all attention paid to even the most seemingly trivial details that you can end up with an advance over the commonly known possible.
Cheers,
- Status
- Not open for further replies.