That was the method used in education here for many years. Potatoes, lemons ... a copper and a zinc pin and then let a clock run. Quite hands on if you ask me. Today the level has gone down considerably.
Have you ever seen a thread here where claimed difference in sound was accompanied by measurements 😉
As far as I can understand, a resistor having a non-linear current versus voltage characteristic is Non-Ohmic. Such a resistor, causes AC signals to undergo changes which involve the creation of new frequecies. A little fiddling with trigonometric identities should illustrate this fact. It should also show that linearly behaving resistors do NOT add anything of their own, apart from obeying V = I*R.
Mathematically, multiplying a series of sines and cosines by a constant, the mathematical equivalent of a signal, DOES NOT change the original distribution of frequencies and their relative amplitudes. In other words, linearly behaving devices cannot change the nature of signals.
A good book about the theory of electronics should explain why linearly behaving components are so important. Their importance lies in their ability to allow electronic engineers to predict what these devices do to signals and to also be confident that these components do not add anything of their own.
Linearity is so important, that it is also used to describe the behaviour of transistors notwithstanding these are extremely non-linear. In fact, the aim of small signal analysis is to take only a very small portion of a transistor's input curve, so that, the transistor's behaviour, is to a very good approximation linear. Differential pairs use small signals which make them behave as if they were linear devices. There are also emitter followers. As can be seen, linearity is at the heart of electronic design. Any reputable book about electronics theory, which does not hide its mathematical basis from readers, should give a good and in depth explanation as to why linearity is central.
Mathematically, multiplying a series of sines and cosines by a constant, the mathematical equivalent of a signal, DOES NOT change the original distribution of frequencies and their relative amplitudes. In other words, linearly behaving devices cannot change the nature of signals.
A good book about the theory of electronics should explain why linearly behaving components are so important. Their importance lies in their ability to allow electronic engineers to predict what these devices do to signals and to also be confident that these components do not add anything of their own.
Linearity is so important, that it is also used to describe the behaviour of transistors notwithstanding these are extremely non-linear. In fact, the aim of small signal analysis is to take only a very small portion of a transistor's input curve, so that, the transistor's behaviour, is to a very good approximation linear. Differential pairs use small signals which make them behave as if they were linear devices. There are also emitter followers. As can be seen, linearity is at the heart of electronic design. Any reputable book about electronics theory, which does not hide its mathematical basis from readers, should give a good and in depth explanation as to why linearity is central.
The claim and the measurement was not on this site. Of course there are plenty of other sites as well. E.g. most professional reviews have both measurements and listening evaluations.
Well, the temp and voltage coefficients of a resistor do actually change the resistance value of the resistor.
There is no non-linear voltage versus current characteristic in this context.
Higher temp raises the resistance, but it's still V/I.
So Ohms' Law is still observed.
Jan
Yes.
There IS one way thermal variation CAN distort a signal, so justifying the mantra "resistors distort", the devil is in the details.
* voltage must be high enough to heat up resistor significantly.
* thermal coefficient must be high enough to produce significant distortion, say 1% or not much lower.
0.1% sensitive golden ears are in the wrong building, Guinness Records office is across the street.
So a few ppm do not apply either.
* this is VERY important: resistor thermal time constant (which is related to its mass) must be short/fast enough for it to vary significantly within the waveform duration ... or it won´t distort it, but simply change level instead, what Jan Didden said above.
Since typical resistor thermal time constant (just measure how long it takes to heat up or cool down) is measured in seconds at least, no variation is possible within the duration of any audible waveform.
Even less at high frequencies, where it might justify "shrillness" 🙄
There IS one way thermal variation CAN distort a signal, so justifying the mantra "resistors distort", the devil is in the details.
* voltage must be high enough to heat up resistor significantly.
* thermal coefficient must be high enough to produce significant distortion, say 1% or not much lower.
0.1% sensitive golden ears are in the wrong building, Guinness Records office is across the street.
So a few ppm do not apply either.
* this is VERY important: resistor thermal time constant (which is related to its mass) must be short/fast enough for it to vary significantly within the waveform duration ... or it won´t distort it, but simply change level instead, what Jan Didden said above.
Since typical resistor thermal time constant (just measure how long it takes to heat up or cool down) is measured in seconds at least, no variation is possible within the duration of any audible waveform.
Even less at high frequencies, where it might justify "shrillness" 🙄
A resistance value that varies with respect to the applied voltage is non-linear in nature. The reason is, for low voltages the gradient of its V-I graph is a value, while at voltages above a certain value, the gradient is different. This change in gradient value "bends" the V-I locus of points, hence, it is a non-linearity.
Regarding, resistance changes due to a resistor's finite thermal capacity, you are right. Energy dissipation in a resistor raises the temperature, but temperature has to increase at a sufficient rate during the life of at least the lowest harmonics.
Regarding, resistance changes due to a resistor's finite thermal capacity, you are right. Energy dissipation in a resistor raises the temperature, but temperature has to increase at a sufficient rate during the life of at least the lowest harmonics.
A resistor that heats itself up and changes value does not comply with the superposition principle and is therefore non-linear. The same holds for any resistor that produces thermal noise and any resistor that goes up in flames when you apply a 1 GW signal to it.
Linearity is a mathematical abstraction. Practical devices deviate from it, the question is how, how much and whether that matters for their application.
Linearity is a mathematical abstraction. Practical devices deviate from it, the question is how, how much and whether that matters for their application.
Last edited:
To all readers: just wait till the screws of the chassis are discussed.
I´d search around before discarding that possibility 🙄
I have seen lots of arguments around iron/aluminum/copper/plated iron (with different materials) chassis so screws would be the next logical step 🙂
I know .. suppose you have electrons boiling off a hot element pinging nearby high voltage elements, would that be dreaded tube sound 😎
All day I just couldn't wait to get home, power up my sound syatem, and listen to my new resistors!
What if measurements are top-notch. For a pre-amp let's say.
And someone ends-up buying or building that stuff, convinced by reading the perfect lab reports.
And lets assume that he even payed a large sum of money too.
Let's also assume that when he plays music on it, he finds out contrary to his own expectations that the single-ended, tube equivalent he used for years sounds subjectively better to him.
Subjectively, to his own set of ears.
And he eventually ends up selling the top lab performing device, keeping his 'lowly' tube gear which his ears do prefer.
Is that problematic or somehow unwarranted?
Because I find it hard to grasp, that someone buys/builds an audio device to enjoy its perfect lab-reports.
I find it more plausible that eventually, the priority may be to please his own set of ears.
That does not imply that what he prefers, with his own set of ears, is what others might prefer.
It is a subjective matter.
It puzzles me that such a simple fact should be denied or ridiculed.
And someone ends-up buying or building that stuff, convinced by reading the perfect lab reports.
And lets assume that he even payed a large sum of money too.
Let's also assume that when he plays music on it, he finds out contrary to his own expectations that the single-ended, tube equivalent he used for years sounds subjectively better to him.
Subjectively, to his own set of ears.
And he eventually ends up selling the top lab performing device, keeping his 'lowly' tube gear which his ears do prefer.
Is that problematic or somehow unwarranted?
Because I find it hard to grasp, that someone buys/builds an audio device to enjoy its perfect lab-reports.
I find it more plausible that eventually, the priority may be to please his own set of ears.
That does not imply that what he prefers, with his own set of ears, is what others might prefer.
It is a subjective matter.
It puzzles me that such a simple fact should be denied or ridiculed.
Last edited:
What has all this got to do with this thread and the OP? And what is the fact to be denied or ridiculed?
- Status
- Not open for further replies.