Great description, Jim. You've answered some nagging questions that have
bugged me for a long time.
Paul, W9AC
Sent from my iPhone5
On Aug 28, 2013, at 8:03 PM, "Jim Garland" <4cx250b@miamioh.edu> wrote:
> Sorry guys, but I can't resist weighing in on this topic one more time.
> Variations on this thread (hole vs. electron conduction, direction of
> current flow, etc.) pop up occasionally on the reflector and to me are
> always interesting, so I hope you'll forgive me for pontificating a bit.
>
> In science, there are often several ways to skin a cat. The various ways are
> always equivalent (if the science is correct!), but their usefulness depends
> on the application. Take something very simple, for example, such as the
> reflection of light from a pane of glass. If you shine a beam of light onto
> a pane of clear thin glass, roughly 4 percent of the light will be reflected
> back toward the source and 96 percent will be transmitted through the pane,
> no matter how clear the glass is. This effect has been known for hundreds of
> years. In the 19th century, this effect was explained by using the language
> of waves, with words like diffraction, interference, transmission,
> reflectivity, index of refraction, and so forth. That language is similar to
> what we use today in discussiing antennas and transmission lines. The
> explanation works great.
> In the early 20th century, however, it was discovered that light
> isn't a wave but a particle (photon). Ditto for radio waves. You can prove
> that light is a particle by reducing the intensity of a light beam and using
> a photodector until you resolve individual photons. They arrive one at a
> time, each one causing a click on your photodector when it arrives.. You can
> do the same trick with radio waves, though it's harder. This was a shocking
> discovery, because once you know that light is made of particles, and that
> 4% of them are reflected, then you've got a real puzzle on your hands. If
> all photons are the same, then how do 4% of them decide whether or not to be
> reflected? It's as if 4% of the photons say to themselves, "Hey, I think
> I'll reflect off the glass instead of zipping through it." But if we have
> identical photons, all interacting in exactly the same way with the pane of
> glass, then all of them should behave exactly the same way. But that isn't
> what happens. This puzzle drove people crazy in the early 20th century.
> But then quantum mechanics came in and saved the day. When you work
> out the details, it turns out that you can analyze light reflecting off a
> pane of glass using the mathematics of quantum mechannics,which treats a
> light beam as made up of particles, which we know it is. When you do this,
> you get exactly the same answer as you do when you use the language of
> reflection, transmission, refraction, interference, and so forth. You can
> also use quantum mechanics to analyze radio waves on a transmission line, or
> radiated off an antenna. But, we'd be crazy to do this. It's so much easier
> to use our conventional wave explanation, even though we know there are
> limits to how well that explanation applies to all situations. Today, we
> know that the wave description of light and radio waves is exactly
> equivalent to the quantum mechanical description for certain situations, but
> it's not as generally valid.The quantum mecanical description is
> horrendously complicated by comparison, even though it is always valid in
> all situations.
> Here's a more straightforward example. Consider a simple capacitor.
> It is convenient to say that a capacitor passes RF current, with a reactance
> that is inversely proportional to the frequency of the current. (Engineers
> often approximate capacitors as short circuits and inductors as open
> circuits, when looking at their impact on very fast pulses.) In all circuits
> involving capacitors, therefore, we can treat the capacitor as an object
> that passes RF current. It's a great way to describe the effect of a
> capacitor on a circuit. However, in truth, no current actually goes through
> a capacitor. There is no charge at all that flows through the capacitor
> (assuming the capacitor is lossless). Instead, we know that the capacitor
> merely stores charge on its plates, and the charging and discharging process
> acts the same as if current flows through.it. The two descriptions are
> equivalent. We can "pretend" that RF current flows through the capacitor and
> we get the correct answer, even though we know it really doesn't flow
> through it at all. In the same way, we can pretend that light or RF
> radiation is a wave, even though we know it's a particle. The wave
> desciption is much more convenient and easy to use than the more "correct"
> particle description.
> And the same situation applies to electron transport in a vacuum
> tube or hole transport in a semiconductor junction. Whatever works and is
> easiest to use is the best way to analyze the situation. In a cathode ray
> tube, electrons go from the cathode to a fluorescent screen and are
> deflected by a magnetic field. We would be nuts to try and articulate that
> process in terms of current flowing from the fluorescent screen to the
> cathode, although in principle we could do so. Similarly, in a vacuum tube,
> if you're interested in how the space charge builds up, it's best to think
> about negatively charged electrons. In a P-channel MOSFET, or a particle
> accelerator (which accelerates protons), you'd best think about holes and
> positive charges.
> Now here's the key point. Whatever description we choose, we have to
> be self consistent, or eventually we'll get hopelessly confused. So even
> though we know plate current in a vacuum tube is carried by negative
> electrons, electrical engineers always speak of the current in an operating
> vacuum tube circuit as flowing from the plate to the cathode, i.e., from
> positive to negative. They do that because the actual sign of the charged
> particles (electrons) doesn't matter from the point of view of the circuit
> explanation. If they described the current in the vacuum tube as flowing
> from the cathode to the plate, then to be consistent they would have to
> describe the current from the power supply as flowing into the positive
> terminal of the power supply.. They could do that, and it wouldn't
> technically be wrong, but it would be confusing, because circuit diagrams
> are drawn with a sign convention in which current flows from a region of
> high potential to a region of low potential. The little arrow in an NPN or
> PNP transistor symbol points toward the direction of current flow (positive
> to negative) as does the arrow in a diode symbol. It's just a convention,
> and it could been chosen differently, but it's a universally accepted
> convention. If someone bucks the convention, and does everything backwards,
> it's not that they're wrong, but they're going to have a lot of trouble
> communicating with other people without causing confusion. So the
> convention, accepted by virtually all electrical engineers and scientists in
> the world, is that current flows from positive to negative, no matter what
> particular particle (proton, electron, hole, ion) carries the current.
> Remember that electric currrent is a statistical quantity, like wind
> velocity, or temperature, or entropy, or specific heat, that only makes
> sense as the average of a large number of individual particle motions. Being
> statistical means electric current cannot be used to analyze the sign or
> direction of individual particles, just as temperature can't be used to
> describe the motion of individual air molecules. So while it may seem
> non-intuitive, individual electrons in a vacuum tube flow from the cathode
> to the plate, but the universal convention is that current in a vacuum tube
> flows the other.way. Sorry, but that's the way it is.
> 73,
> Jim Garland W8ZR
>
>
> _______________________________________________
> Amps mailing list
> Amps@contesting.com
> http://lists.contesting.com/mailman/listinfo/amps
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|