A lot of electronics questions

Factual Questions
1

I have a pretty basic understanding of electronics. I know that resistors resist the flow of electrons. Capacitors store energy in an electric field, inductors store energy in a magnetic field, diodes allow current to flow in one direction, transistors are used for amplification or as switches, etc. I also know how to analyze circuits like this with differential equations or phasors. However, I can’t ever put this into practical use. Like if I see a schematic I can kind of understand how it works, but I certainly couldn’t have come up with the design by myself. For my own edification I am trying to understand how data transmission works – both wired and wireless.

1. For instance, start with wired data transmission. Suppose I am transmitting a digital signal. I have a wire in California and another one 3,000 miles away in Florida. I am transmitting binary 101 at a frequency of 1 Hz to Florida. Does this accurately represent the potential of the wire at a given time?

Time = 0 sec - CA 000000000000000000000 FL (intially nothing being sent)
Time = 0.1 sec – CA 11111111111111111111 FL
Time = 0.9999 . . . 9 – CA 1111111111111111 FL (just before switching)

This is where I am confused. I am not sure if the potential of the line changes instantaneously
Time = 1 sec – CA 000000000000000000000 FL (entire wire instantaneously changes potential)
or
Time = 1 sec – CA 000000000000 . . 11111111 FL (signal propagates through wire at speed of light)

Where in the second instance there is some brief period of time where closer towards Florida the line is a logic 1 and closer to California the line is a logic 0. Part of me is leaning towards it being the second because the signal will travel at the speed of light there is some finite, albeit very small time, where the logic 0 hasn’t made it (3000 miles/670 616 629 mph * 3600 seconds/hour = ~16.1ms for the signal to travel from California to Florida) to Florida. But I have always heard that the potential of a conductor is constant.

Does that make any sense?

2. My next question relates to bandwidth. We all have heard about how fiber optics have allowed for much higher data transmission rates. Why, or more specifically, how? I know that some optical fibers allow for dense wavelength division multiplexing (basically frequency division multiplexing using colors of light), but why does this allow for more data to be sent? You can’t transmit the data any faster since through copper since you can already send it at the speed of light. You can’t send any more data since you can send multiple signals via frequency division multiplexing in both copper and fiber. It seems, to me, that the only difference between the two would be in the amount of loss or electromagnetic interference each transmission medium would have or what physical device in California is used to put the signal on the line. Can they put it on the fiber more quickly than on copper which would allow for more bandwidth?

3. I have cable and DSL internet coming into my home. So I can only have 1 wire coming into my house, the internet and cable share a line. I have a 10 Mhz splitter coming into my house to divide the two signals up and send one to my cable modem and the other to my TV. How does the splitter know which signal is which? I assume that digital signals are coming into the splitter, bifurcated, and then sent to two filters which I assume is just a low pass and a high pass filter each allowing a certain frequency spectrum through. But I don’t understand how this would work in real life.

4. I can’t seem to wrap my head around how antennas work for wireless transmission. Suppose I wanted to create an FM transmitter. Basically you would need a microphone to get the person’s voice. Some kind of amplifier to increase the voltage level. Then something to modulate the signal onto a carrier signal in the FM frequency band. Then this signal has to be transmitted through the air via an antenna. How would you get this to work? The way I understand it, any electrical current passing through a wire creates a magnetic field. Is an antenna basically a wire used to create a magnetic field that another antenna picks up, amplifies, and then plays through a speaker? Wouldn’t this have to be a pretty large current for any significant distance? Seems to me that there would be tons of electromagnetic interference with all the electronic devices around. How would the antenna shape or size effect this? I imagine you would need a lot of power to transmit audio from a radio station or from outer space.
I see schematics and they have inductors here and capacitors there, a transformer or two, etc. Besides the basics of capacitors help to keep voltage steady and reduce noise, inductors can keep current steady via the magnetic field, and transformers scale a voltage up or down with a corresponding effect on the current to keep power about the same, I have no idea when to do what.

I hope I used the correct terminology, and that this is at least somewhat coherent. Can anyone recommend a book with real life, practical examples that would explain a lot of this? I have taken several electrical engineering courses, but frankly, I can analyze the things using phasors, differential equations, or PSPICE, but I couldn’t do anything electrical in the real world to save my life. I am a software guy anyway so I guess it doesn’t really matter in the grand scheme of things, but I would still like to know.

Please let me know if you need clarification. My brain is pretty scattered sometimes, and I have problems articulating my thoughts.

2

Oh, and related to the book question, if anyone knows of any good (i.e., active) message boards that discuss electronics that would be even better. I have found the best way to learn something is to surround yourself with people who know what you want to learn.

3

If you’re looking for some good books that will answer your questions, here’s two:
The Art of Electronics

The Arrl Handbook for Radio Communications

If you only get one, I’d get the ARRL handbook, which may be in your local library (if you’re in the US). especially since most of your questions deal with communications, modulation, antennas and transmission lines, etc.

4
  1. The signal propagates at the speed of light, but you’ve also got to figure in the capacitance of the line itself. A real world transmission line isn’t anything like you’d picture a short length of wire to be. We tend to model transmission lines as a series of circuits containing resistance, capacitance, and inductance, like so:
    http://www.microwaves101.com/encyclopedia/images/Transmission%20lines/general_Tline.jpg

The potential of an IDEAL conductor is constant. The potential of a real world conductor isn’t anywhere close to constant. Heck, you can get several volts of potential difference just in your relatively short lengths of house wiring.

Instead of neatly going from 0 to 1, the signal is going to rise kinda towards one, maybe overshoot a bit, then go back to slightly under 1, then over again, etc. until it finally settles on 1, while slowly (at a fraction of the speed of light) propagating down the line.

  1. Capacitance is your major frequency limiter. The longer you make the copper, the more capacitance you have to fight, and the more your higher frequencies get attenuated.

In fiber, the light just kinda bounces down through the fiber, so you get much less attenuation. You can therefore pump much higher frequencies accurately through the fiber.

  1. Cable internet and DSL internet aren’t the same thing. Cable internet uses your TV cable. DSL generally uses your phone line. DSL is much higher frequency than your phone line, so is easily switched off using a simple filter.

For a cable modem, the splitter doesn’t know which signals go to which. It sends both signals in both directions. Your TV knows what to do with the TV signals. It doesn’t know how to handle the internet signals, so it just ignores them. Likewise for your cable modem. It only looks at the signals that it knows what to do with.

  1. For FM, you need a voice signal, a voltage controlled oscillator (VCO), an amplifier, and an antenna. If you shape your antenna in such a way that it’s tuned to the amplifier’s output frequency, then most of the energy from the amplifier will radiate out as radio waves. If the antenna isn’t tuned to the right frequency, then a lot of the energy will reflect off of the antenna and go back into the amplifier, possibly damaging it. It’s therefore much more important to have your transmitter antenna well tuned than your receive antenna.

RF stuff is kinda voodoo. It’s difficult to really understand, and you can’t develop a good hands on idea of how it works unless you really play around with it a bit. The ARRL handbook at your local library is a good place to start if you want to do some reading.

Figuring out what part to use where just takes experience. When I first started out designing things, I did a lot of “how the heck am I going to do that” :eek: but after several years, that feeling starts to go away. You can read all you want, but you can’t develop design skills without designing. Start with simple things. Play around with things in a simulator. It’s a lot cheaper than buying parts. On the other hand, you do occasionally need to buy parts so you can figure out what you did wrong when they don’t work. Troubleshooting is also a learned skill.

If you don’t have it already, run out to ye ol local bookstore and buy a copy of “The Art of Electronics” by Horowitz and Hill. It’s a darn good book.

5

I see that Arjuna34 and I seem to like the same books.

6

Compare the different DSL technologies (specifically, look at how ADSL breaks up the spectrum for Tx/Rx) in the box at the right of this article: 25 upstream bins, 220+ downstream, and one big bin for voice transmit/receive down at the bottom. Also check out the article on Wavelength-division multiplexing where it states that “modern systems can handle up to 160 signals and can thus expand a basic 10 Gbit/s* fibre system to a theoretical total capacity of over 1.6 Tbit/s over a single fibre pair.” Last but not least check out this paragraph on the advantages and disadvantages of copper and fiber optics. It boils down to very low losses/noise on optical fiber, versus the lower material cost of copper. If copper meets your needs for bandwidth and noise, go with it.

(* - note that fibers are very very skinny, so you’ve got many parallel channels each carrying different info in the time domain as well as in different frequency domains, vice one big fat copper slab)

That’s exactly how it works**. You dedicate the internet signal to one set of frequencies and place correctly-tuned notch filters on the line. You dedicate the other signal (TV? on your phone line?) to some other pack of frequencies, and use another notch filter, or a high/low-pass filter if the ends of the spectrum don’t matter. Just realize that upstream, someone is doing frequency-division multiplexing to fit both signals on the line; you need to do the de-muxing in order to get useful information out of the combined wave. If you know someone who has an older ADSL provider, you’ll notice that all of their phone jacks that aren’t used for internet have a filter in-line with the phone. If you want, you can crack one of these open and look. I haven’t done it, but I don’t imagine there’s much more than a low-pass filter inside to let 0-4 kHz through: a resistor, a capacitor, maybe an op-amp, and some epoxy.

** - for DSL over phone lines, not for cable internet over co-ax. engineer_comp_geek explained the co-ax case very well.

I don’t know enough antenna theory to help you with this – I was a MechE in college, not an EE.

For the fourth question, you could probably get a study guide for a Ham Radio license and find most of the basic antenna theory you’d need.

7

A few clarifications:

On the first part, it sort of depends on what you mean by ‘ideal’ conductor. Usually (and as you did here, engineer_comp_geek), it means ‘lossless’. However, the potential being everywhere the same only describes the steady state, or when you’re dealing with static fields. Even in a perfect conductor, a signal can only propagate at the speed of light at maximum. Unless you’re ignoring the speed of light entirely (which you rarely can if you’re talking about signal transmission). Also, signals in real-world wires don’t all propagate at the speed of light, either.

Another thing about signal transmission - you may also have reflections at both the receiving and sending end (which would become more important at higher frequencies). With a constantly changing data transmission, the actual waveform is likely to be some superposition of pulses, which won’t necessarily be 0 or 1 at any given point (but most likely will be interpretable as 0 or 1 at the end, hence one of the major advantages of digital communications).

This line right here makes me wonder if you’re slipping into a common misconception about electricity - that it is the flow of electrons. If you are, disabuse yourself of this notion quickly. Electricity is the flow of energy (electromagnetic energy, to be precise). It can be transmitted via charge carriers (which are electrons in almost all cases) but as wireless technology or that big bright burning thing in the sky shows, this is not a necessary part.

This might help you a bit in understanding how wireless signal transmission works. It is not massively different from sending signals down a wire; in this case it’s just an electromagnetic wave traveling through air (or whatever else). An obvious complication is that very high frequencies have to be used in order to prevent the losses that would occur if you tried to send the signal straight out. (Tied up with that is that you’d need a massive antenna as well).

As for books, AOE is a good start, if a bit severely out of date on some parts[sup]*[/sup]). Once you’ve learned a bit (or a lot) more, if you’re really serious about learning more about signal transmission, give one of Howard Johnson’s books a try.

[sup]*[/sup]It provides far more practical and ‘current’ examples of usage than other textbooks, which means it has aged quickly. Probably the majority of it is still relevant, just don’t take all of it as gospel.

8

Regarding RF, Maxwell’s equations explains the details pretty well if you can follow the math. However it’s possible to get a general idea of how RF transmission works if you keep in mind two electromagnetic concepts.

First, you are correct that an electric current passing thru a wire creates a magnetic field. For a straight wire, the magnetic field is wrapped around the wire in a cylinder, and its strength depends on the intensity of the electric current and (of course) your distance from the wire itself.

The key point in creating an electro-magnetic wave is that the current in the antenna wire is not constant DC: An AC current in the wire forces the magnetic field around the wire to vary in strength (not to mention flipping direction with each half of the AC cycle).

This varying magnetic field is detected at a remote antenna because of Faraday’s law. In its original form this law states that a change in magnetic flux thru a closed loop of wire (e.g. thrusting a magnet into the loop) will cause a current to flow in the loop. The varying magnetic field from the source antenna serves the same purpose; it is the mechanism that couples the current in the source antenna with the reacting current in the receiver.

There are two obvious questions for this interpretation: (1) The source antenna is an unterminated wire; how do I force any current into it, since there’s no place for the current to go? (2) Though there are some loop antennas in use, many receiver antennas are single wires; how does Faraday’s law apply?

To address (1), the frequency of the driving current is made high enough that the signal’s corresponding wavelength is comparable in size to the length of the antenna itself. Obviously no current can be flowing at the very tip of the wire antenna, but the charge along the wire can be “sloshed” back and forth if the driving voltage changes quickly enough. If the antenna is the right length (usually a multiple of 1/4 the wavelength of the driving signal), the antenna actually resonates this motion, much like the way a pipe organ tube can be cut to resonate at one specific sound frequency.

That last comment about resonance also helps explain at least one way to cut thru the EM clutter. For a single-wire antenna, the length of the wire determines the wavelengths at which the antenna resonates, and hence the wavelengths that are most easily transmitted/received. More complicated geometries plus the use of arrays of antennas allow the broadcaster to take even greater advantage of this property–measured variously as antenna Q or antenna gain–but there are plenty of other factors (impedance matching and tuning circuits are perhaps even more important, especially on the receiver).

The answer to (2) is a bit more complicated, but the final explanation is similar to what’s going on in (1). For his eponymous law, Faraday only needed the “loop” of wire in his experiments because he was working with relatively slow-changing magnetic fields (you can only move a magnet so fast by hand thru a wire coil), and he was detecting these flux changes using a DC meter. The broadcast magnetic field, on the other hand, moves quickly enough that it can also “slosh” the electrons in the receiver antenna–even though the antenna isn’t in a loop form needed to complete any DC circuit.

Again, Maxwell’s equations explain this far more elegantly that the lumbering post above, and I’ve completely omitted detail about how the electric field couples antennas (I’ve always though the “changing magnetic field” explanation is more intuitive, even if it leaves out a lot of detail). Still, I hope it helps…

9

Antennas are mysterious. It just plain turns out to be a fact of life that if you have alternating current in a wire you can detect alternating electric and magnetic fields in the space surrounding the wire. The interaction of the two fields results in an electromagnetic wave that travels away from the wire at the speed of light. And here again, that’s just a fact of life. Each field falls off as the distance and the power, which is their product, falls off as the square of the distance.

The antenna currents of high powered transmitters can be pretty big. For example, a 50 kW transmitter with an antenna having a 70 Ohm input impedance has an antenna current of √(50000/70) = 27 A.

Noise is reduced by several methods. Static is reduced by making the receiver frequency selective, a directional receiving antenna, and using frequency modulation. Internal noise from the random motion of charges resulting from heat is reduced by using low noise first stages in the amplifier and in extreme cases cooling the first stage.

There are various fancy error correction, or detecting, codes available in digital transmission. Someone else will have to address that because it’s out of my area.

Further explanations are available if there are questions.

10

I haven’t had time to read everyone’s posts, but I will be back tomorrow evening after work. Thanks for the quick replies!

11

All of the devices you mention are useful for AC circuits, and it is critical to remember that AC operates at a frequency. Sometimes a circuit is directly driven by a powerful AC signal (anything you plug into the wall in the US is being driven by a 110v signal at 60 Hz). Other times an engineer must consider sources with frequency varying over a range (e.g. audio from a microphone), and sometimes he/she must consider the frequencv of unwanted “noise”, particularly high-frequency RF noise.

For AC, the properties of capacitors and inductors are such that an engineer can treat them like resistors whose resistance varies with the applied frequency (yes, these devices also affect the phase of the AC signal, but let’s not get too far ahead of ourselves). To summarize briefly, inductors provide more impedance to a signal at higher frequencies, a capacitor provides less (and, counterintuitively, larger capacitance actually introduces a smaller impedance. A basic capacitor is two very-close-but-not-touching plates; increase the size of the plates or move them even closer together, the capacitance goes up, and intuitively it’s easy to see how either of these changes would make it “easier” for an AC signal to cross the separating dielectric).

For example, often in RF applications, you’ll see an inductor called a “choke” or “RFC” placed on the DC line near the positive terminal of whatever is providing DC voltage. This choke acts as a resistor to block any high-frequency RF from getting into the power supply, where it could be damaging (the battery doesn’t care; inductors look no different from a straight wire to DC). Another example is a isolation capacitor placed on a line going to a speaker: This capacitor blocks DC 100% from going to the speaker (a fast short to ground that could overheat the coil), but if the capacitor is large enough it will pass all signals above a certain threshold freqauency (hopefully the low end of you audio range, ~40Hz) easily.

The key point in all this is that AC introduces variation that these new components take advantage of, and frequencies of the AC you are working with heavily influences this.

12

An electrical signal in a conductor?

13

It propagates at a substantial percentage of the speed of light in a vacuum. See velocity factor.

14

The signal, as mks57 points out, propagates at a substantial percentage of the speed of light. But do not confuse this with the speed of the electric current itself–the “drift velocity” of the electrons, which is actually quite small (Wiki sez A 20-gauge copper wire carrying 5A moves electrons at a speed of 1 mm/sec). Sort of like pushing a long pole; the pole itself doesn’t have to move much, but once you start pushing at one end, the other end begins to move almost instantly.

15

What is a signal, if not current? Drift velocity is not the same as the speed of current.

Once again, electric current is not the flow of electrons. It is the flow of energy.

16

>I see that Arjuna34 and I seem to like the same books.

EVERYBODY and you seem to like the same books. Those are two incredibly fantastic books. Though, for my money, I’d buy “The Art…” first if I had to start somewhere.

For reasons woven pretty deep into the fabric of the universe, electrons all repel one another everywhere. And, any kind of change at all can only propagate at a speed of c (often called the “speed of light” but it’s much more fundamental and important than light is). These two facts alone, as they intertwine in various ways, cause much of what you study in electronics.

17

Current is defined as the flow of electrical charges. One Ampere is equal to one coulomb of charge per second passing a given point. There is an unGodly number of electrons in even a small piece of a copper wire so the electrons need not move very fast to have that amount of charge passing a point. For example an electron has a charge of 1.610[sup]-19[/sup] Coulombs. It thus takes 6.2510[sup]18[/sup] electrons to make one Coulomb of negative charge. Copper has an atomic weight of 63.5 so that number of grams contains Avogadro’s Number (6.02*10[sup]23[/sup]) of copper atoms. If they have one free electron per atom it only takes 0.6 milligrams of copper to contain one Coulomb of moveable charge.

18

I’ve been loving this thread - I’m in the same boat as the OP and have been tangentially wondering these things for a while. So as to not hijack this thread, I’ll link to my related So I want to do this electronics stuff for a living. What’s my next step?

19

:smack: Of course this is right; not sure what I was thinking there.

I’d still say electricity is primarily about the flow of energy, not necessarily electrons or even charge. Current is defined as the flow of charge. The energy propagates at c, the speed of light in a vacuum [And while this is technically always true, it’s almost always less confusing to refer to the signal traveling at less than c].

20

No, it propagates at the speed of light though the medium immediately surrounding the conductor. If this happens to be free space, then it will propagate at c. In air, this is something like 99% c, while for typical plastic insulations, this can vary between about 60-75% c.