John Curl's Blowtorch preamplifier part III

Status
Not open for further replies.
Thermal noise. White.

Excess noise, may be 1/f, may not appear at all. I'm sure I read somewhere that metals don't have significant 1/f noise.

Correct on both. I've learned not to make absolute statements because I've been reading the literature on this for 40yr. For those interested, this is from my link of a day or so ago. Notice the reference thick film metal resistors, white. The experiments here were probably done under fairly rigorous conditions, I expect nothing less.
 

Attachments

  • pic3.jpg
    pic3.jpg
    201.8 KB · Views: 394
Are you going to tell me propagation velocity doesn't change with material in a conductor?

In addition to the normal equation there is the relative permeability. Now copper is .999991 and silver is .99998.

So there should be a different velocity of propagation. Now what the impurities are and their permeability is I suspect no one really knows, but can offer a good guess.

Now just for giggles if a copper wire was formed by a steel die all bets are off as to permeability and uniformity. Good thing steel dies for wire forming are almost extinct.

Wow huge difference between copper and silver, not. And the kicker is (http://web.hep.uiuc.edu/home/serrede/P436/Lecture_Notes/P436_Lect_07.pdf) there is virtually no EM field in a conductor after a few inches ( skin depth) so how does the wire affect propagation? Learn the physics before you make stuff up.
And something I learned, almost all the energy that flows in a conductor is in the magnetic field.
 
Quote:
Originally Posted by simon7000
I think the issue is at higher currents that happens, but at extremely low currents there isn't as much interaction so the scattering may be more.

Are you saying that resistance varies with current? If so, an awful lot of small-signal apparatus would cease to work.

No I am saying that at high current densities the charges repel each other and that results in a more uniform charge flow.

Quote:
I did heat a copper wire to form oxides on the surface to see what effect that had. While hot it showed lots more energy in the target frequency band. Once it cooled it was the same.

How did you heat it? Was it all at exactly the same temperature? If not, have you heard of the thermoelectric effect?

I used a propane torch as what I was interested in was "microdiodes" formed in the oxide. Not surprisingly no effect was found.

Quote:
This is where we don't communicate. 100 ohms at room temperature will show 1.3 nanovolts per root hertz.

Thermal noise. White.
There is no argument here.

Quote:
Why you think this will change with frequency, I don't get.
Excess noise, may be 1/f, may not appear at all. I'm sure I read somewhere that metals don't have significant 1/f noise.

Quote:
So my take is that in an imperfect conductor the signal would take different paths with different velocities of propagation hence slightly different arrival times and levels thus at the end of the conductor the effect would be to appear to have some noise added. When two sine waves are passed I would expect this process to yield a bit of the sum and difference signals that should also appear as noise at those frequencies.

My take is that at higher current levels things average out faster and appear smoother.

You need to define what you mean by 'velocity of propagation'. Then maybe someone will try to explain it to you yet again.


It takes some finite length of time for a signal to pass through a conductor. In a homogeneous conductor, the signal should travel uniformly. In one with contamination or one that is not homogeneous I expect the signal to take all paths in proportion to the paths' conductance. This will result in some small part of the signal taking more or less time. Yes this is probably in picoseconds or even less time for a typical audio cable.


You need to think about orders of magnitude and timescales. Ordinary scattering produces resistance, which is linear over a huge range of current densities and also constant over a huge range of frequency. Resistance produces thermal noise. To see your postulated effect you will need to look well below thermal noise, to see the noise modulated by the signal. Thermal noise in a wire is tiny, so your postulated signal will be much smaller than tiny. My guess is that experimental error will be much greater than the effect you are looking for, so the effect will be invisible even if it exists.

I am fairly certain your take is wrong. I am fairly certain that your experiments will not see the effect you are looking for. I am fairly certain that your experiments will see something, but then others will tell you what you are seeing: ordinary experimental errors.

I am seeing things just above the noise level. I do not see any significant difference in my original tests of different solder types. I did see larger differences in connectors and a bit of difference in wire types. Low quality wire showed more band of interest energy than oxygen free copper and that more than silver wire. This seems to agree with what is reasonable and expected.

This was not a one off test. I did a number of different tests and versions of the gizmo. If I redo it there are a number of changes I will make.
 
JN,

Perhaps we disagree on this. When a charge enters a conductor one does leave. In a typical interconnect this may take 30 nanoseconds or so. At which point an equilibrium is reached.
Make up your mind..is it 30 nanoseconds, or is it picoseconds???

A charge doesn't just enter and wait for a brethren to get off the bus at the other end.
Electrons don't simply crowd the bus at higher currents, dislike their crowded conditions, and suddenly repel one another. If that were the case, heating a bar of copper, all the electrons would jump off.

On some forums, making this stuff up might get over. But not here.

Here, you have to make sense..

As I've stated, your test setup is terribly uncontrolled. You play until you find something that seems to agree with ridiculous notions, then present us with little snippets.

You seem to be afraid to actually show us what you've done, we might find the fatal flaws. I've told you where your confounders lie, you ignore it.

Jn
 
If I may be so rude as to offer a picture, closing the switch initiates both currents instantaneously and the line maintains constant charge. Or maybe someone could tell me how one end of the line knows how to wait for the other? BTW it does not matter at all if this is an open line transmission line or a circular loop, in the later case a 3D field simulator would help with the tedious maths.
 

Attachments

  • lpic.jpg
    lpic.jpg
    23.8 KB · Views: 259
Last edited:
30 nanoseconds to get from one end of the interconnect to the other. Variation of pico seconds if for example we were sending a 0 to x volt square wave.

If it were instantaneous that would be a bit faster than the speed of light.

Same thing if you shine a flashlight at a wall from the time you turn it on until you see the spot of light it takes some time.
 
If it were instantaneous that would be a bit faster than the speed of light.

Please refer to the picture, you disagree that the current at each end of the battery starts at the same instant (throwing the switch) so tell me which end waits 30ns and how does it know? I could post the solution for an open wire transmission line which is a current staircase to infinity which straddles exactly the solution for the lumped L at low frequencies but that would probably be a waste of time.
 
Last edited:
Please refer to the picture, you disagree that the current at each end of the battery starts at the same instant (throwing the switch) so tell me which end waits 30ns and how does it know? I could post the solution for an open wire transmission line which is a current staircase to infinity which straddles exactly the solution for the lumped L at low frequencies but that would probably be a waste of time.
There was a young lady of Wight,
Who traveled much faster than light,
She departed one day,
In a relative way,
And arrived on the previous night.
 
Please refer to the picture, you disagree that the current at each end of the battery starts at the same instant (throwing the switch) so tell me which end waits 30ns and how does it know? I could post the solution for an open wire transmission line which is a current staircase to infinity which straddles exactly the solution for the lumped L at low frequencies but that would probably be a waste of time.

Charge travels from negative to positive.

How does a TDR work, particularly on an open cable?
 
simon7000 said:
No I am saying that at high current densities the charges repel each other and that results in a more uniform charge flow.
Why would the charges repel each other more at high current densities? There are exactly the same number of charges at high current and zero current, moving in exactly the same crystal lattice. Do you know anything about conduction in metals?

It takes some finite length of time for a signal to pass through a conductor.
What do you mean by "pass through"?

Perhaps we disagree on this.
Yes. He is right. You are wrong.

You seem to be inventing your own stories about how wires work, then setting up experiments based on these stories, then using these stories to misinterpret the raw data you see on your instruments. Science progresses by each generation 'standing on the shoulders of giants', not by each generation starting from the ground again. In order to advance knowledge, you first need to learn what is already known.
 
So, if I understand correctly, if one closes that switch and then (before the charge could reach the other end of the conductor [ns's...us's]) opens it, then that charge is captured by the conductor, reportingly doing this will act as, some kind of, a charge-pump.

Question, what is the maximum charge for 10Mtrs of RG58? :)

P.s. with the switch at the negative pole. With the switch at the positive pole one will (presumably) create a charge-vacuum.

P.s. Question, what is the maximum vacuum that 10Mtrs of RG58 can withstand? :)
 
Last edited:
Status
Not open for further replies.