W65C02S RDY / BE pins & first build advice

[BigEd]

I suppose the first thing to say is that all advice, if sincerely given, reflects the habits and experience of the person giving it. My feeling is that distraction is a potential problem, which is one reason to try to apply some discipline to learning how to do something. Another reason to apply discipline is to build knowledge and understanding in stages. Trying to learn everything all at once seems to me like it will make it more difficult. Iteration and repetition helps, too: aiming to learn everything from a single ambitious project seems to me very... ambitious. A series of projects, of increasing sophistication and difficulty, seems a better idea.

As for limiting yourself to 50 things a year, I'm on record as estimating that I manage about one productive weekend a year. People can be wildly different, but I'd be impressed to see 12 good projects a year from anyone.

But, as they say these days, you do you, and hopefully you'll get somewhere you want to get to, and also enjoy the journey.

[Osric]

Paganini wrote:

Osric wrote:

Like a lot of folks, as a teenager I thought of electricity like “water in a pipe”. But I think this analogy is fundamentally unhelpful because when you apply pressure (“voltage”) at one end of the pipe it is easiest to think of it as instantaneously appearing at the other end of the pipe (assuming the pipe is full of water). Perhaps if I had thought of it as “empty pipes” that would have been a better mental model.

That's also how I learned the basics in 4H club. I also think it's a bad analogy. Water runs downhill. The explanations and illustrative diagrams always confused me because I couldn't figure out how gravity fit into the picture. Nowadays I think of it as *air pressure.* This is not necessarily better, I guess, but at least it makes sense to me!

I guess if you want to think of electrical potential like gravity, it’s maybe worth looking at as a function of position. I think https://www.youtube.com/watch?v=j3GrOKre__0 does a good job of advocating this point of view.

Paganini wrote:

One thing that might help, maybe, is to notice that there is not really any such thing as a "signal" moving in the wires. The idea of "signal" is an abstraction that lets us think about logic instead of electrons. Similarly, "voltage" is not a thing that moves around. Voltage is a measurement taken by us, the observers, that describes the relationship between two points in a circuit.

I guess we have different definitions or we disagree on both of these points. To me a “signal” has a clear definition as the measure of voltage or current over time. You can see a signal on your oscilloscope; you can “amplify” it with a transistor; you can send it through space with an antenna. The bits we send from a CPU to a RAM component are signals; in digital we exclusively concern ourselves with “logic signals” and try to define when the signal has to be read to make our imaginary clean world of 1s and 0s, but signals are not an abstraction, they are the underlying voltage levels and they are messy when things aren’t working. Similarly, “voltage” is a measurement of a signal value at a particular point in time between two particular locations; it’s what we plot on our oscilloscopes. And it does take time to move around, because we are essentially measuring the force being applied by an electric field and the electric field builds up in the wires and in the space between them wires at something approaching the speed of light - so it’s moving, albeit very, very quickly. For steady state circuits we can pretend it’s everywhere like you describe but for fast transient circuits — like our computers running GHz speed — we can’t.

Paganini wrote:

What we are measuring, specifically, is the number of free electrons (or "electron holes" that free electrons could move into).

If you take this view, why don’t the electrons from the positive terminal of a battery flow into the negative terminal if I connect two batteries but don’t close the circuit? If there are “holes” in the negative terminal, and “free electrons” in the positive terminal, they should flow without a circuit…as indeed they do when you charge a comb on your hair and then touch it to a neutral object which is first attracted to the static and later repelled when the charge equalizes. But connecting two batteries in series does not ruin them.

Paganini wrote:

In the classical model, because Ben Franklin was a bit of a goober, the point that has more "electron holes" is positively charged, while the point that has more free electrons is negatively charged. If you connect the two points with a conductor, the free electrons will move from the negative point to fill in the holes at the positive point. Classical current is the movement of the "electron holes" in a conductor, but what is really happening is that electrons are zipping around in the *opposite direction.*

I think that’s a bit unfair to Franklin, who was 100 years ahead of electron theory and had to make some arbitrary choices about which particles moved and which were stationary. What’s a bit more surprising to me is that he had the notion that charge moved only in one direction (which I’m not sure is strictly true if we think about ions in solution subjected to a potential).

Paganini wrote:

Ideally, all points in a circuit that are electrically common have the same voltage (0V) when compared to each other. In reality, though, electrons move very quickly, but *not instantaneously.*

Actually I think it’s clearer to say that electromagnetic fields propagate quickly. The electrons themselves cause this propagation of the field by moving tiny distances; they do not “zip around” the circuit.

Paganini wrote:

So, it takes the tiniest bit of time for the battery to "suck" the electrons out of the circuit and charge up points B and C. How long this is depends on how big the wire is, what it's made of, and what else is going on around it.

In fact, the thing that matters is the distance between the point where the voltage is applied and the point where you’re interested in seeing if the electrons are moving, and what kind of dielectric fills that space. The wires dominate in channeling the fields to the destination, but there could be short cuts that don’t involve the wires and are problematic.

Paganini wrote:

But more importantly, to change the voltage at the 6502's clock input pin some electrons must move. Those electrons come from somewhere! Where? The answer is ground.

Eventualy the answer is ground. But initially, the electrons that are already there can just move around slightly to create the electric potential. This will happen whether there is a circuit or not, I believe.

Paganini wrote:

The ground side of a circuit is like a free electron reservoir. If some electrons go out from the clock input of the 6502 into the clock output of the oscillator, then the same amount of electrons must also go into the 6502 through its ground pin. This means that, temporarily, the 6502's local ground pin is slightly more positive than the rest of the ground side of the circuit. Meanwhile, the oscillator has a few extra free electrons that it just received from the 6502. It has to send those electrons back to the 6502 so that the free electron reservoir can settle down. If you don't suck, you can't blow. This is the signal return path. Unfortunately (in some ways) for us, electrons like to take the path of least resistance. If the loop back between the ground pins of the oscillator and 6502 is long or windy those electrons will try to find some other way to go, which can include through other ICs that might happen to be in the way. Havoc may ensue.

I think the return path is via the path of least inductance but I have a weak understanding of this point. I also don’t think it is the circulation of current that causes the ground levels not to match. Are you sure you are correct in your explanation?

Paganini wrote:

Moreover, as the oscillator switches voltage at its output pin and the 6502 follows almost (but not quite) instantaneously at its input pin, tiny amounts of current will be constantly flowing back and forth between those two pins, and also between the two ground pins of those ICs. Even though our (logic) signal is (abstractly) traveling in one direction (from oscillator to 6502), we are actually dealing with alternating current! We potentially have to think about capacitance, inductance, impedance and all kinds of complicated math. Because of electromagnetic fields we might experience intermittent glitches on a signal caused by complex relationships between switching patterns of *other signals.* This is why Ed keeps talking about rules of thumb. To debug these sorts of problems you need (often very expensive) equipment, and a real command of electronics math. This is something that most hobbyists (including me!) would rather avoid.

Finally, I guess it's worth pointing out that even the classical model is not "what's really going on." As Ed says, you can go all the way down to quantum mechanics, where electrons are not "things" that "move" but are more like ripples in fields. Every model is a description in some sort of language, more or less precise, of what we observe. There is no "what's really going on" that we can access independently from our perceptions. At some point you have to stop looking for the next turtle down and start building computers.

I think for my purposes getting as far as “electrons are tiny magnets that create electric and magnetic fields” will be sufficiently low-level in the stack of turtles, with possibly a tiny splash of relativity to understand how the charges create magnetic forces. I’m OK with doing the math to understand it at this level; but I don’t need to fully grok the new model as long as my partially constructed model can explain the problems I’m having at any given time. Because at the end, I do agree with you: at some point, I want to be building things, not studying

[BigDumbDinosaur]

Osric wrote:

BigEd wrote:

- level 0, put things together and see if it works, then change things until it does, or until you run out of patience...Trying to learn everything all at once seems to me like it will make it more difficult...

I guess I plan to do exactly what you recommend against and I should spend a bit of time understanding your objection/what the disadvantages are.

I am bowing out of any further discussion of this topic, as it has become clear to me you are resistant to expending a modicum of time to study basic electronics and understand foundational relationships.  As Ed intimates, knowledge of the physics themselves isn’t a prerequisite to being a successful hobbyist.  What is a prerequisite is understanding the principles, which is where I think you should focus your efforts right now.

That said, I will leave you with this parting thought.  Swimming against the tide results in the expenditure of a lot of energy but gains you little headway.

[Osric]

BigDumbDinosaur wrote:

Osric wrote:

BigEd wrote:

- level 0, put things together and see if it works, then change things until it does, or until you run out of patience...

I guess I plan to do exactly what you recommend against and I should spend a bit of time understanding your objection/what the disadvantages are.

I am bowing out of any further discussion of this topic, as it has become clear to me you are resistant to expending a modicum of time to study basic electronics and understand foundational relationships that any electronics designer must know in order to be successful.  I will leave you with this thought.  Swimming against the tide results in the expenditure of a lot of energy but gains you little headway.

I’m really sorry to lose your insight and maybe at some time point you’ll change your mind and chime in again.

I’ll reiterate that I do not think I amt resistant to learning the theory. I just want to focus my learning on deeply understanding theory that is in my way at any given moment, rather than gaining a surface understanding of a broad swath of ideas that will enable me to build faster but without true understanding of what the causes the issues that I get to avoid by following best practice without understanding why it is the best practice.

[GARTHWILSON]

You'll notice I have not gotten into the math here.  One of several reasons is that it's not really necessary for what we're doing here.  As long as we have a good concept of what goes on and what the effects are, we'll avoid the problems that can keep our stuff from working.  I imagine we all agree that the more we understand, the better, but that there's not any magic threshold we need to reach to enjoy making stuff that works.

Osric, I'm having a bit of a hard time latching solidly enough onto your question and request for confirmation in your post on the last page, and while I've been thinking about it, so many more posts have been added and it's hard to keep up; so this might be out of order, and certainly incomplete, as new material gets posted faster than I can address it.

I don't know if I've see the exact "famous veritasium video" you mention, but I know I've at least seen related ones, and they regard the transmission-line effect of the parallel conductors.

The electric field comes from the voltage.  There doesn't have to be any current to form this field, although if it's across a capacitor, you'll need some current to charge up the capacitor to some voltage, just as it takes some movement to store energy in a spring.

The magnetic field comes from the current, regardless of voltage, although to get the current going in an inductor, you'll need some voltage across it, just as it takes some torque to get a flywheel going.  (It also takes torque in the opposite direction to slow it down and to stop it.)

You've probably seen the inside of a mechanical wristwatch, where the escapement which sets the speed has a little flywheel with a spring on it.  The spring, like a capacitor, and the flywheel, like an inductor, together resonate at the (hopefully) desired rate, going back and forth, so the ticking makes the watch keep accurate time.  At the two parts of the cycle where the flywheel is stopped, all the energy is stored in the spring; but at the two parts of the cycle where the flywheel is turning its fastest, the spring is at its point where is has no tension on it.  They're not in phase with each other.  The analogy in the computer is that this resonance is desirable for the clock oscillator to go some controlled number of MHz (although the frequency is normally governed by a crystal which emulates an extremely tight LC circuit).  Other things that ring and cause problems will generally be much, much higher in frequency, out of control.

A difference however is that as the current in a wire changes, and the magnetic field which goes around it therefore changes, this field, as it goes out and in, as it builds and collapses, respectively, can cut through nearby wires, and try to collapse onto them, making corresponding current changes in them in the opposite direction, so the net current change is reduced or eliminated, which is why twisted pairs and transmission lines are so much better behaved for getting signals through.  The magnetic-field lines will be perpendicular to the electric-field lines.  I have sketches illustrating mutual inductance at viewtopic.php?p=55094#p55094 .

The water-pipe analogy only goes so far, for various reasons, and I can't think of any way to make it extend to mutual inductance.

Quote:

3- This is where I get confused; we start to talk about impedance in the ground connections of the originator. I do not understand this part. My best guess based on what I’ve tried to understand is to imagine the originator is driving a bus, like a data bus, and sending 0xFF down the bus - now it’s switching 8 outputs at once. On either end (rising edge or trailing edge) this is going to cause some ground bounce, but it seems like the ground bounce is really only going to cause us a problem on trailing edges because there’s less margin in our logic signals on the low side than the high side: and when the signals go from 0xFF to 0x00, the inductance between the transmitting chip’s ground reference and the system ground cause issues.

Many CMOS inputs and outputs are more or less symmetrical, unlike TTL.  WDC's output drivers can pull up all the way to Vcc, and pull up just as hard as down, and many 5V CMOS parts' inputs have their thresholds at 2.5V.  Jeff Laughton has an excellent post (actually topic) about thresholds at viewtopic.php?f=4&t=6594 .

Paganini wrote:

That's also how I learned the basics in 4H club. I also think it's a bad analogy. Water runs downhill. The explanations and illustrative diagrams always confused me because I couldn't figure out how gravity fit into the picture. Nowadays I think of it as *air pressure.* This is not necessarily better, I guess, but at least it makes sense to me!

The "running downhill" part is true when it's out in the open, like a river, not bound by a pipe.  In the pipe situation, if you have something like a U-shaped pipe, and pour water into one end, it will go down the U, traverse the bottom, and up the other side, trying to equalize the pressure on the two sides; so in that case, yes, it will go uphill.  Even when air is not bound, cold air will run downhill, like I experience while riding bike in the local canyons in cold weather, and there are these V's in the hills on the uphill side with no sun on them, and cold air in those V's runs down and goes across the road where they come down to it, such that when I go by there, it's suddenly very cold.

Quote:

Those electrons come from somewhere! Where? The answer is ground. <snip>

Note however that conductors don't particularly need to 'get' electrons from somewhere.  Copper has only one electron in its outer valence shell, so it's very mobile, and atoms can pass electrons like a ring of performers on a stage all passing hats in sync from one head to the next, where they all start with one hat and no matter how long the passing goes on, they each always have one hat, or between landings, are passing one hat to the head in front of them while the performer behind is passing another hat to their head, all to the rhythm of the music.  I think there was an example in the Chitty Chitty Bang Bang movie when Dick Van Dyke was doing "The Old Bamoo" at the fair, the performance that got him the money to buy the old car.  Traffic signal sensors, if their sensitivity is adjusted per a particular federal standard (which unfortunately many are not), can sense a bicycle wheel.  The sensor is a coil tucked into a slot in the pavement, and the controller puts an alternating current through it, and senses changes in the complex impedance.  It does not "weigh you;" in fact, it's not necessary to even touch it.  My rims are aluminum, able to conduct electricity around in a closed circle, and when I put the wheel right on the wire (not across it, nor inside the loop), the traffic signal controller senses that I'm there waiting for a green light.  The wheel didn't have to receive electrons, but instead, the ones already there just started going around in a circle, collapsing the field set up by the sensor loop in the pavement.

Quote:

Unfortunately (in some ways) for us, electrons like to take the path of least resistance. If the loop back between the ground pins of the oscillator and 6502 is long or windy those electrons will try to find some other way to go, which can include through other ICs that might happen to be in the way.

A nice thing about a ground plane is that because of mutual inductance, the return path will go through the ground plane, taking on the shape of the trace carrying the signal, because that's easier for it than taking the shortest path, and it eliminates some problems.

Quote:

Actually I think it’s clearer to say that electromagnetic fields propagate quickly. The electrons themselves cause this propagation of the field by moving tiny distances; they do not “zip around” the circuit.

True; in the analogy of water in a pipe, if you press some water into one end, the effect is seen at the other end very soon after, even if all the water in the pipe moved only a fraction of a percent of the length of the pipe. The electrons may drift very slowly through a wire, but the phase propagation velocity is basically the speed of light if in free space, or usually more than half the speed of light in a transmission line, depending on the dielectric constant.  Without looking it up to review, I think the propagation velocity through FR4  PC board is about 2/3 the speed of light.

[Osric]

GARTHWILSON wrote:

Osric, I'm having a bit of a hard time latching solidly enough onto your question and request for confirmation in your post on the last page, and while I've been thinking about it, so many more posts have been added and it's hard to keep up; so this might be out of order, and certainly incomplete, as new material gets posted faster than I can address it.

Sorry for not being clear! I'll try to distill my question(s) into true/false questions so that they're easier to answer both in terms of time required and in terms of figuring out what I'm trying to ask. My goal in asking these questions is to confirm my understanding or shine a light on my misunderstandings, not to put people on the forum to the trouble of writing tomes in an attempt to educate me (heck, in your case you've already written and published your tome, it's totally unreasonable to expect more!). I do appreciate people's attempts to explain, too, of course, but I know it's very time consuming: please feel free to provide one-word true/false answers (or no answer at all).

In any event the questions from that post in true/false form are:

1. True or false: The electric field represents the force applied to charges in space, but it takes time to set up the electric field surrounding any conductor, and the time required depends on the “inductance”.
2. True or false: This setup time exists whether or not the circuit is open or closed
3. True or false: The electric field charges the wire to the potential that we’ve applied if the circuit is open
4. True or false: The electric field causes current to flow if the circuit is closed

If the answer to any of the above is "false", I need to go back and re-review the content that makes me believe these things to identify whether I have misunderstood it or if the content is wrong.

GARTHWILSON wrote:

The electric field comes from the voltage.  There doesn't have to be any current to form this field, although if it's across a capacitor, you'll need some current to charge up the capacitor to some voltage, just as it takes some movement to store energy in a spring.

Great, this would appear to correspond with my question 3 and imply that it is true, validating my understanding.

GARTHWILSON wrote:

The magnetic field comes from the current, regardless of voltage, although to get the current going in an inductor, you'll need some voltage across it, just as it takes some torque to get a flywheel going.  (It also takes torque in the opposite direction to slow it down and to stop it.)

This appears to indicate I have a misunderstanding. I thought that when we connect a voltage source to an open wire, the electric and magnetic fields would get established over time. Then if we remove the voltage, the magnetic field would collapse. However based on what you're saying now I think that the magnetic field collapses once the fields settle down: that is, we connect voltage to the wire, the electric and magnetic fields get set up down the length of the wire at for the sake of argument 0.6c, bounce around a little and then settle to a 0 current situation where the electric potential is now available at all points on the wire but there is no longer a magnetic field. When the circuit is closed and current flows, the magnetic field will set up along the length of the wire. It seems to me that this setup must occur "backwards", that is the voltage source on a long wire sets up an electric field that propagates from the voltage source along the length of the wire to its open end, but when we close the circuit the magnetic field will propagate from the open end back to the voltage source as some of the free electrons shuffle toward the end of the wire.

[True or false?]: As a separate matter, it is clear to me that a capacitor acts like a short circuit turning into an ever stronger resistor causing current to flow while its electric field gets set up, while an inductor (any wire or coil) acts as a open circuit turning into an ever less resistant resistor while its electric field gets set up. That'd better be true or I really don't understand

GARTHWILSON wrote:

The water-pipe analogy only goes so far, for various reasons, and I can't think of any way to make it extend to mutual inductance.

Fill the water pipe with bucky balls? ... indeed, I have completely abandoned the water pipe way of thinking about it as misleading and wrong.

GARTHWILSON wrote:

Many CMOS inputs and outputs are more or less symmetrical, unlike TTL.  WDC's output drivers can pull up all the way to Vcc, and pull up just as hard as down, and many 5V CMOS parts' inputs have their thresholds at 2.5V.  Jeff Laughton has an excellent post (actually topic) about thresholds at viewtopic.php?f=4&t=6594 .

I'll have to review my data sheets and maybe be more careful about not using a mishmash of whatever parts I have in my parts drawers.

GARTHWILSON wrote:

A nice thing about a ground plane is that because of mutual inductance, the return path will go through the ground plane, taking on the shape of the trace carrying the signal, because that's easier for it than taking the shortest path, and it eliminates some problems.

This seems to correspond to what I was trying to say/ask of Paganini, that it's the path of least inductance not the past of least resistance that guides the current on the return path. Or maybe another way to say it is that as the fields in the ground plane need time to get set up, the field will set up closest to the signal trace first and so current will flow right underneath the signal trace initially, perhaps spreading into a wider and wider flow over time, but if the current flow is short lived fundamentally this should result in the current flowing the same path as the trace because there is no time for the field to spread out over the ground plane, and this is not the path of least resistance, which could be directly across some shorter straight line in the ground plane, but a path determined by the mutual inductance ... which is why I questioned Paganini's point.

[BigEd]

In case it's at all helpful, when I used to maintain a hydraulic model in my head, I would imagine a balloon attached to the pipe for a capacitor to ground. Or, a stretchy rubber membrane halving a hollow sphere, for a capacitor which joins two signals.

For inductance, imagine some kind of turbine or water wheel in the flow, perhaps with a flywheel attached. It will resist changes in flow, and store energy in its rotation, and deliver it back. For inductance between signals, couple two turbines to the same flywheel.

I don't really bother with a hydraulic model any more. And, although I do have a maths degree, I've never used equations beyond Ohm's Law and Kirchoff's Law. Never studied or memorised Maxwell's. I have sometimes considered graphs - for example, for transistor characteristics.

Although my approach to the universe in general is to apply physics, when it comes to hobby electronics, I think in terms of simple switch-level models, and with rules of thumb otherwise. (See the topic "Techniques for reliable high-speed digital circuits" for my amateur take.)

The one thing I picked up early on from formal training is the idea of single-clock synchronous design, whereby the clock is the only signal where edges matter. And the time immediately before and after the active clock edge is the only time which matters. This is very powerful, has served me well, and suits HDL design styles very well too.

[Paganini]

Osric wrote:

I guess we have different definitions or we disagree on both of these points. To me a “signal” has a clear definition as the measure of voltage or current over time. You can see a signal on your oscilloscope; you can “amplify” it with a transistor; you can send it through space with an antenna. The bits we send from a CPU to a RAM component are signals; in digital we exclusively concern ourselves with “logic signals” and try to define when the signal has to be read to make our imaginary clean world of 1s and 0s, but signals are not an abstraction, they are the underlying voltage levels and they are messy when things aren’t working.

Well, what you say here is true, so far as it goes. But a signal is like a number: it's an *idea.* A signal has no physical reality itself: it's a way of thinking about a physical reality that lets us do something useful.

Quote:

Similarly, “voltage” is a measurement of a signal value at a particular point in time between two particular locations; it’s what we plot on our oscilloscopes.

This is wrong, though. Voltage is simply a measure of potential between two points. There is voltage between your feet and your carpet, and in the wires going to your toaster. No signals there! What you see on your oscilloscope is whatever you set it up to measure. You could measure frequency (think radio) and still be looking at a signal. Or you could measure current. I'm under the impression that old logic families represented signals with current, not voltage, but I am not strong on that.

Quote:

Paganini wrote:

What we are measuring, specifically, is the number of free electrons (or "electron holes" that free electrons could move into).

If you take this view, why don’t the electrons from the positive terminal of a battery flow into the negative terminal if I connect two batteries but don’t close the circuit?

They might! But remember, we don't know anything about how many free electrons (or electron holes) are at the negative (positive) terminal of the batteries. All we know is that, when comparing each battery's *own terminals* there is a 9V potential between them. When you put two batteries together as you describe, there might be a *brief* current flow, like charging a capacitor.

Quote:

Actually I think it’s clearer to say that electromagnetic fields propagate quickly. The electrons themselves cause this propagation of the field by moving tiny distances; they do not “zip around” the circuit.

They do, in fact. Electrons move very very quickly. How quickly depends on a lot of things (some of them already mentioned, like the size of the wire and what it is made of). I think what you're referring to is that an individual electron does not travel a very long distance. For the "last" electron in the circuit to move into the place of the "first" electron, all the electrons in between have to ripple one "step," like marbles in a pipe, which happens almost instantly, while the individual electrons displace almost no distance. (There's a cool video where someone actually made a "marbles in a pipe" circuit to demonstrate the principle, but I'm having trouble finding it!) So, our signals travel around at nearly the speed of light, the electrons themselves move at millions of miles per hour, but are only displaced in space a few inches or yards.

Quote:

Paganini wrote:

So, it takes the tiniest bit of time for the battery to "suck" the electrons out of the circuit and charge up points B and C. How long this is depends on how big the wire is, what it's made of, and what else is going on around it.

In fact, the thing that matters is the distance between the point where the voltage is applied and the point where you’re interested in seeing if the electrons are moving, and what kind of dielectric fills that space. The wires dominate in channeling the fields to the destination, but there could be short cuts that don’t involve the wires and are problematic.

Good point; I meant to include length in my list. It is maybe not *the* thing that matters, bit is *a* thing that matters. In the Primer, Garth reiterates over and over that a good practice to avoid many problems is to keep your wires short!

Quote:

Paganini wrote:

This is the signal return path. Unfortunately (in some ways) for us, electrons like to take the path of least resistance. If the loop back between the ground pins of the oscillator and 6502 is long or windy those electrons will try to find some other way to go, which can include through other ICs that might happen to be in the way. Havoc may ensue.

I think the return path is via the path of least inductance but I have a weak understanding of this point. I also don’t think it is the circulation of current that causes the ground levels not to match. Are you sure you are correct in your explanation?

Yes, but maybe I could have stated it better. Electrons do take the path of least resistance; this is basic electricity. In DC circuits there is no reactance to worry about. However, as I described, when our ICs are switching very quickly they become AC circuits, and the return path is the path of least *impedance.* Impedance is the AC equivalent of resistance. A ways back you expressed confusion about impedance at the ground connection of a signal originator. The point I was trying to get across, badly I guess, is that your ground return network needs to be thoughtfully designed because if the connection between ground pins of the sender and the receiver is not the path of least impedance, the signal return path will find other ways to go, including through intervening ICs.

Quote:

I think for my purposes getting as far as “electrons are tiny magnets that create electric and magnetic fields” will be sufficiently low-level in the stack of turtles, with possibly a tiny splash of relativity to understand how the charges create magnetic forces. I’m OK with doing the math to understand it at this level; but I don’t need to fully grok the new model as long as my partially constructed model can explain the problems I’m having at any given time. Because at the end, I do agree with you: at some point, I want to be building things, not studying

Do you know about the website "All About Circuits?" There is a kind of (free!) online textbook there that covers the basics of AC, DC, and explains terms like resistance, inductance, reactance, and impedance. It's what I used when I first got started in the hobby. It was enough to help me establish some "rules of thumb" like Ed describes so that I have some stable hobby boards (and some not so stable ones! )

[Osric]

Paganini wrote:

Osric wrote:

Quote:

Paganini wrote:

What we are measuring, specifically, is the number of free electrons (or "electron holes" that free electrons could move into).

If you take this view, why don’t the electrons from the positive terminal of a battery flow into the negative terminal if I connect two batteries but don’t close the circuit?

They might! But remember, we don't know anything about how many free electrons (or electron holes) are at the negative (positive) terminal of the batteries. All we know is that, when comparing each battery's *own terminals* there is a 9V potential between them. When you put two batteries together as you describe, there might be a *brief* current flow, like charging a capacitor.

We're closer than I thought on what a signal is, though we still disagree a bit; a signal has a physical reality, and the numbers need interpretation to have meaning - even the voltage between my feet and my carpet might have meaning if I know how to interpret those numbers in a useful way ... but let it pass as close enough.

We are not close on this one though: if voltage is caused by holes, and you connect two batteries in series then all the holes should get filled by the first battery. This is how it works with static electricity; we can create a charged item and tap it to a neutral item and the charge spreads between them; the neutral item will then be repelled by the charged item. I contend it is not how it works in the case of a battery, which must have a connection back to its other terminal in order to enable the electrons to circulate in any meaningful number. A capacitor I contend is a totally different animal again, where the electric field holds the energy. Edit: I can see a way I have this wrong, even for the "holes" model. You have to argue that the holes on battery 1 keep the electrons from filling the holes in battery 2 because they are pulling on the electrons and holding them in the battery. You can argue that since the holes in battery 1 are present, there are no "free electrons" on the side with the surplus electrons because they are accounted for by the as yet unfilled holes. Therefore no current can move until something is providing electrons into the holes on the other side of the battery, hence the need for a complete circuit.

Paganini wrote:

Quote:

Actually I think it’s clearer to say that electromagnetic fields propagate quickly. The electrons themselves cause this propagation of the field by moving tiny distances; they do not “zip around” the circuit.

They do, in fact. Electrons move very very quickly. How quickly depends on a lot of things (some of them already mentioned, like the size of the wire and what it is made of). I think what you're referring to is that an individual electron does not travel a very long distance. For the "last" electron in the circuit to move into the place of the "first" electron, all the electrons in between have to ripple one "step," like marbles in a pipe, which happens almost instantly, while the individual electrons displace almost no distance. (There's a cool video where someone actually made a "marbles in a pipe" circuit to demonstrate the principle, but I'm having trouble finding it!) So, our signals travel around at nearly the speed of light, the electrons themselves move at millions of miles per hour, but are only displaced in space a few inches or yards.

It's the drift velocity I am referring to which we seem to agree is quite slow, so perhaps we agree. At the end of the day the charges on the surface of the conductors inch along slowly. The fact that they are moving at all is what happens quickly, which is the electric field propagating. Perhaps we agree, I can't quite tell. I don't find your description to match my mental model, but I may just be misreading it - the speed of transfer of power is based on the propagation of the fields, not the movement of electrons.

Paganini wrote:

Quote:

Paganini wrote:

So, it takes the tiniest bit of time for the battery to "suck" the electrons out of the circuit and charge up points B and C. How long this is depends on how big the wire is, what it's made of, and what else is going on around it.

In fact, the thing that matters is the distance between the point where the voltage is applied and the point where you’re interested in seeing if the electrons are moving, and what kind of dielectric fills that space. The wires dominate in channeling the fields to the destination, but there could be short cuts that don’t involve the wires and are problematic.

Good point; I meant to include length in my list. It is maybe not *the* thing that matters, bit is *a* thing that matters. In the Primer, Garth reiterates over and over that a good practice to avoid many problems is to keep your wires short!

I'm not sure if we are agreeing here or not; here I am not talking about the length of the wire, but the distance between where the voltage is applied and any other wire. So the length of the signal wire isn't what I was trying to emphasize here, but the fact that as current starts to flow down wire 1, all adjacent wires will feel the influence of the electromagnetic field from wire 1. You'd originally said "if you don't suck you can't blow" or something similar, but this is more like feeling the movement of air that's adjacent to you and moving a bit yourself, even if you're not in that circuit of sucking/blowing.

Paganini wrote:

Paganini wrote:

Quote:

Paganini wrote:

This is the signal return path. Unfortunately (in some ways) for us, electrons like to take the path of least resistance. If the loop back between the ground pins of the oscillator and 6502 is long or windy those electrons will try to find some other way to go, which can include through other ICs that might happen to be in the way. Havoc may ensue.

I think the return path is via the path of least inductance but I have a weak understanding of this point. I also don’t think it is the circulation of current that causes the ground levels not to match. Are you sure you are correct in your explanation?

Yes, but maybe I could have stated it better. Electrons do take the path of least resistance; this is basic electricity. In DC circuits there is no reactance to worry about. However, as I described, when our ICs are switching very quickly they become AC circuits, and the return path is the path of least *impedance.* Impedance is the AC equivalent of resistance. A ways back you expressed confusion about impedance at the ground connection of a signal originator. The point I was trying to get across, badly I guess, is that your ground return network needs to be thoughtfully designed because if the connection between ground pins of the sender and the receiver is not the path of least impedance, the signal return path will find other ways to go, including through intervening ICs.

I'm not sure if what I said is correct or what you're saying is correct - inductance is part of impedance but not the whole story, and I still think this might be the past of least inductance rather than impedance. At least to me, it makes sense that the shortest path through a ground plane impedance wise ought to be the straight line path between the endpoints of the trace, rather than the potentially winding path taken by the trace on a different layer of the PCB.

Paganini wrote:

Do you know about the website "All About Circuits?" There is a kind of (free!) online textbook there that covers the basics of AC, DC, and explains terms like resistance, inductance, reactance, and impedance. It's what I used when I first got started in the hobby. It was enough to help me establish some "rules of thumb" like Ed describes so that I have some stable hobby boards (and some not so stable ones! )

I tried to take a look but was a bit put off by the home page. Is there a direct link to the online textbook you mean?

[GARTHWILSON]

Osric wrote:

Calculus seems a lot easier to me in my old age than it did as a student in university, so dealing with the math is not really a hurdle anymore.

So I'm taking the liberty to get into just a little bit of math here, for the math major.  I apologize that this is all text (so far).  I can imagine a lot of pictures that would bring some of this stuff down to earth.  Hopefully a few word picture sprinkled in will help anyway.

Osric wrote:

[True or false?]: As a separate matter, it is clear to me that a capacitor acts like a short circuit turning into an ever stronger resistor causing current to flow while its electric field gets set up,

The voltage across a capacitor is proportional to the integral of the current through it; so if you start a zero charge on it, then suddenly connect it to a circuit with a voltage, the current will be limited only by whatever impedance that circuit has.
farads = coulombs / volts
So for example if you put 1 coulomb of charge into a capacitor and it charges it up to 2 volts, it has half a farad, or 500,000µF.
coulombs = amps * seconds; so:
farads = amp-seconds / volts  (or you can put "micro" in front of "farads" and "amp-seconds," or put "nano" in front of them both, etc., for example µF and µA-seconds, or pF and mA-nanoseconds, or other combination that makes the powers-of-a-thousand prefixes divide or multiply correctly)
...and of course you can work that around with simple algebra, so the situation where you initially connect the capacitor is:
volts = amp-seconds / farads

Depending on the circuit however, the charging current, and therefore the rate at which the voltage on the capacitor changes, usually drops as the capacitor's voltage gets closer to the final voltage.  The time constant ("TC") of a resistor and capacitor in series is a convenient amount of time needed to get the capacitor to 1-1/e times the final voltage, e being the natural-log base, approximately 2.718 (or, if you like rational numbers for scaled-integer math, 193/71 is .001% high, and 23225/8544, which is still representable with 16-bit integers, is only .00000025% high).  Anyway, 1-1/e is about 0.63, and if you start at 0V and final voltage is 5V, that's 3.16V which in CMOS circuits is close to the minimum often-guaranteed solid logic-'1' level of 3.5V.  2 TC gets you to .865 of the final voltage, or 4.32V in this case, and so on.  For many applications, 5 TC is considered fully charged, although of course you approach "fully charged" asymptotically, so theoretically you never quite reach it.

Quote:

while an inductor (any wire or coil) acts as a open circuit turning into an ever less resistant resistor while its electric field gets set up. That'd better be true or I really don't understand

The current through an inductor is proportional to the integral of the voltage across it, which is the opposite of the situation with a capacitor, so initially the current is zero, and it increases.  The voltage across it is proportional to the first derivative of the current through it.
henries = volts * seconds / amps
So for example if 1V applied for 1s increases the current through the inductor by 2A, that coil has ½H of inductance.
...and of course you can work that around too, and use the prefixes as above to get us into nanohenries, nanoseconds, milliamps, whatever.

In numbers that are more realistic for our applications, suppose an input has 5pF of capacitance.  If we could get rid of all unwanted parasitic reactive elements and just feed it an instantaneous change in voltage from a 100Ω source, 1 TC would be 5E-12 * 100Ω = 500ps, or ½ns, 1ns for 2 TC.  Unfortunately there are parasitic elements all over the place, and can be hard to model them all mathematically, or measure them, particularly with equipment we hobbyists can afford, which is why it's good to just have an understanding of what this stuff does and then just do our best "good engineering practice" (there's that nebulous term again!) to stay far from the gray zones and out of trouble.

Note that while resistance consumes power, inductance and capacitance don't; they only store energy, and can give it back later, or in the case of mutual inductance, it may just be transferring the energy to an adjacent conductor.  If you put a sine wave into a resistor, the current will be in phase with the voltage, and the power lost to heat, in watts, is simply volts times amps.  If you apply the sine wave across the inductor, the current will be 90° behind the voltage, and if you apply it to a capacitor, the current will be 90° ahead of the voltage.  In practice, there are no absolutely perfect components, but you get the idea.  "Resistance" and "impedance" then are not the same thing in AC circuits.  Impedance is a complex number, meaning it goes on a graph and not a number line.  In the frequency domain, the X value is the "resistance" part, and the Y value is the "reactance" part.  Mathematicians use i for the Y part, but in electronics, I usually stands for "current," so we use j instead, like 1260Ω-j400Ω.  The complex impedance could be represented as X,Y, or as magnitude and angle.  In the digital stuff we would also have to deal with the time domain, and could get into things like convolution integrals and Laplace transforms.

All the parasitic elements in the circuit interact with each other.  It's not possible to eliminate them, but they can be minimized, or in some cases damped.  When they're not, the ringing, "splashing," etc. can turn one's circuit flaky or just plain non-op.  If you've ever bounced a ball in a hallway where all surfaces were reflective, you've probably heard the buzzing effect which does not not represent the normal sound of the ball bouncing.  Or perhaps you've been in a large, multi-deck concrete parking garage where all surfaces were acoustically reflective, and you're far from a car whose burglar alarm was sounding "honk-honk-honk..." and with all the echoing, it sounded like one continuous honk, but as you got closer, the honks began to separate and you could tell is was not actually continual.  The analogy relates to how we want to design our boards.  When it's not practical to control or even measure these parasitic circuit elements, we can stay out of trouble by making the circuit relatively compact to keep traces short, and have at least a relatively fine grid of ground connections if not an actual ground plane, and a power-supply decoupling capacitor at each IC.  If you can also do a relatively fine Vcc grid too, all the better.

Quote:

GARTHWILSON wrote:

A nice thing about a ground plane is that because of mutual inductance, the return path will go through the ground plane, taking on the shape of the trace carrying the signal, because that's easier for it than taking the shortest path, and it eliminates some problems.

This seems to correspond to what I was trying to say/ask of Paganini, that it's the path of least inductance not the past of least resistance that guides the current on the return path. Or maybe another way to say it is that as the fields in the ground plane need time to get set up, the field will set up closest to the signal trace first and so current will flow right underneath the signal trace initially, perhaps spreading into a wider and wider flow over time, but if the current flow is short lived fundamentally this should result in the current flowing the same path as the trace because there is no time for the field to spread out over the ground plane, and this is not the path of least resistance, which could be directly across some shorter straight line in the ground plane, but a path determined by the mutual inductance ... which is why I questioned Paganini's point.

One way to look at it is that the conductors themselves don't really carry the energy, but instead only guide it, and the energy is stored and carried in the space between the conductors, and you have to charge up that area.  Things like twisted pairs or traces running across a ground plane will minimize that area, whereas requiring the return current to go through a ground trace that's far away increases that area, and the return may try to take advantage of closer conductors that the signal should not be coupling into.

Then we get into transmission lines that have a characteristic impedance.  All I'll say about that here is that there's capacitance between the conductors (which is determined mostly by the separation distance, the interface area, and the dielectric constant of the separating material), and every conductor has inductance, and we're dealing with mutual inductance, resulting in the "characteristic impedance" looking into one end of the transmission line, and if the other end has a resistive load that matches it, it looks like the line is infinite in length and the signal does not reflect off the end and ever come back.  I'll try to find a picture of a model of how the distributed capacitance and inductance work.  As the signal goes down the line, it's kind of like holding an empty ice-cube tray at an angle under the fawcet at the kitchen sink, and pouring water into the topmost boxes, and as they fill up and begin to overflow, they start filling the next pair of boxes down, which, when full, start filling the following ones, and so on.

[JimBoyd]

GARTHWILSON wrote:

I don't know if I've see the exact "famous veritasium video" you mention, but I know I've at least seen related ones, and they regard the transmission-line effect of the parallel conductors.

I think the veritasium video in question is one of these:

• The Big Misconception About Electricity
  How Electricity Actually Works

[Paganini]

Osric wrote:

Paganini wrote:

Do you know about the website "All About Circuits?" There is a kind of (free!) online textbook there that covers the basics of AC, DC, and explains terms like resistance, inductance, reactance, and impedance. It's what I used when I first got started in the hobby. It was enough to help me establish some "rules of thumb" like Ed describes so that I have some stable hobby boards (and some not so stable ones! )

I tried to take a look but was a bit put off by the home page. Is there a direct link to the online textbook you mean?

Hmm. I replied to this thread yesterday morning, but the toobz seem to have eaten it! Anyway, here (again) is a direct link to the All About Circuits online textbook: https://www.allaboutcircuits.com/textbook/

And here is a link to an article that seems pertinent to this thread: https://www.allaboutcircuits.com/techni ... impedance/

Also, I remember a video someone posted a while back in which an experimenter had some kind of scanning device pointed at a PCB so you could actually see the current return path. They demonstrated a variety of situations, such as what the current return path does when it encounters a void in the ground plane. It was very cool. I tried googling around for it with no luck. Does anyone remember better than I do where that video came from?

[BigEd]

I think Dave Jones has done this - maybe this video, where he traces a Gigatron built two ways:
EEVblog #1176 - 2 Layer vs 4 Layer PCB EMC TESTED!

Quote:

What difference does a 4 layer PCB make to EMC radiated emissions compared to an identical 2 layer PCB? And why?

Dave does H-Field near-field probe testing on two otherwise identical PCB's.

Edit: Amusingly, it's excerpted here, as Dave himself was searching for it in his own content.

[Paganini]

GARTHWILSON wrote:

If you can also do a relatively fine Vcc grid too, all the better.

Hi Garth! I am curious about this one. We tend to place more emphasis on the quality of the ground return network than on Vcc. Why is that? I was reading Dr. Howard Johnson the other day, and came across this:

Quote:

(1) When the driver switches HI the return current (transient and DC) must re-enter the driver through its power pin. When the driver switches LO the return current must re-enter the driver through its ground pin.

This suggests to me that the Vcc network should be just as important as the GND network. Why isn't it?

[BigEd]

I had the same thought - my answer (to myself) is that the nature of inputs is that they are referenced to ground: how large the voltage is, whether or not it's over some threshold, whether it crosses some threshold.

It might well be possible to make circuits which detect signal vs power rail, but we rarely see those.