Comment by powertower | Hacker News Reader

Comment by powertower | Hacker News Reader

Parent Full thread

> Electrical signals in a copper wire travel at approximately 2/3 the speed of light.

An interesting fact - in those copper wires (or any type of wires) the electron velocity itself is less than 1 centimeter / second (and often in the mm/s range). Whether it's DC or AC, it does not matter much.

Think of the wire as a tube with a bunch of ping-pong balls inside - filling it up tightly from start to end. If I stick my finger into one end of the tube, a ball will come out at the other end almost instantaneously (regardless of how long the tube is), but the balls themselves only moved a small distance - and only at the speed/velocity I pushed my finger at. But the "signal" "traveled" at a much higher velocity.

So in DC - electrons only travel a few centimeters / second (not anywhere near light speed), and in AC - you're pushing/pulling the same electron back and forth for eternity at about the same speed as it travels in DC.

Now it's of course a bit more complicated (the electrons rattle around at much greater speeds, and the mentioned velocity is a net average), and there are some edge cases, but for the most part the above is accurate.

I'm just not 100% sure of the speed of electrons in a superconducting loop (Cooper-pair electrons) - if they are truly relativistic or not.

And if this does not blow your mind, you should not be reading about electricity. PHDs in the electrical field don't even know this stuff.

http://en.wikipedia.org/wiki/Drift_velocity

TL;DR; If you raced a snail and an electron in a wire together, the snail would win - https://www.youtube.com/watch?v=jbi7gJTPSXk

One of the coolest things about superconductors is "kinetic inductance"[1]. There is no energy in a normal current. As you say, the electrons don't travel all that fast most of the time. They feel a field, accelerate until they bump into something and start all over again. That's resistance; the energy stored by an inductor or capacitor (a current or voltage) is in the magnetic or electric field nearby.

This isn't true for Cooper-pairs in superconductors. They go fast enough that there is true energy stored in their velocity; there's nothing to stop them, they're a superfluid. This is used for certain types of detectors (MKIDs), and means that there is a bit of extra L in superconducting transmission lines, above the geometric inductance.

[1] http://en.wikipedia.org/wiki/Kinetic_inductance

It's not really the case normal metal transmission lines don't have some "extra" L, is it? The magnetic field still penetrates into any real metal, so you have some extra inductance.

But I agree that it's very important to understand that a superconductor is not the same thing as a "perfect conductor", and the kinetic inductance is a great example of this. The ideas behind MKIDs are very cool.

What blows my mind is that PhDs don't know this stuff.

I have only high school level ed, but I had to train many CS grads for real-life positions, since they were absolutely not fit for job. High schoolers were grasping the ideas much better.

You know, I thought a lot about this. In India, there is a weird phenomena that I noticed: many of the best teachers want to teach, not college students, but high school students preparing for the rigorous engineering entrance examinations. Of course, it helps that they get paid a lot more than university professors, but the ones that taught me, for instance, had a genuine desire to impart knowledge where it would matter the most. They loved it when we would solve a difficult physics/math problem. By contrast, they complained that most college students were too distracted with other things. I found this to be true when I went to college, when, of a class of about ~500 students, only about 50 were actually honestly learning.

My hypothesis is that many college grads don't make the most of their easy access to learning, but have an incredible sense of self-entitlement. Those in high school are still at a very impressionable age; curious about the way the world works and hungry to find out more.

A PhD gives you a specialty, not a well-rounded education. That's why it exists.

CS PhD students are not electrical engineering PhD students.

Plus, as far as I know this is interesting trivia rather than practical information for most engineers. I have a BSEE and don't remember the speed of electrons ever being brought up; for everything I ever learned it was more important to know about the speed of electricity. (But then I did software my whole career so maybe I'm missing something.)

It was mentioned that the direction of current is opposite the flow of electrons. But again you never cared about the electrons, just the electricity.

EE work is usually done with traditional, well-characterized materials. Treating drift velocity as anything other than part of a constant or a temperature variance equation is pretty rare.

It's more common if you're doing novel research.

So if my tube of ping-pong balls is infinitely long, am I sending my signal infinitely fast?

Nope. Imageine instead of ping-pong balls you had a blue squash ball (TIL: the color is important! [0]). They're squishy, and so when you poke the ball at your end, in reality you're causing it to visibly deform. It then pushes back out to a sphere, moving away from you. In doing so, it pushed into the next ball. This process of deform->reform->...deform->reform takes time, and that time is the propagation delay.

[0]: http://en.wikipedia.org/wiki/Squash_(sport)#Playing_equipmen...

No signal can travel faster than light. If you put pressure sensors along the infinitely long tube, you'll see that the signal reaches the points at different times.

We're just so used to seeing it happen instantaneously that it's slightly difficult to imagine otherwise

The signal would only travel at the speed of sound in ping-pong balls. What that is precisely is very difficult to estimate. Try "wave propagation in granular materials" for a Googleable phrase. Even the case of a perfectly aligned 1-D chain of ideal ping-pong balls is not straightforward to analyze; the confinement force probably has a significant effect. (disclaimer: not a physicist)

You're forgetting about propagation of force within the ping-pong ball. It is not immediate.

For which (TIL) the speed of propogation is limited by the speed of light - except (maybe) gravity: http://www.metaresearch.org/cosmology/speed_of_gravity.asp

Reply on news.ycombinator.com