Electron Flow versus Current Flow

Status
Not open for further replies.

#12

Joined Nov 30, 2010
18,224
Here's some more confusion! I was teaching myself electronics for 4 years before I had a teacher, so I decided that a diode symbol looks like a speaker symbol. The bar represents the magnet and the wide end of the arrow emits electrons like an speaker emits sound.

I also decided that schematics are arranged with positive at the top because it makes the math easier. When I can't figure out something by using conventional flow, I revert to electron flow because electrons can't lie.

Confused enough? Just pick one and be consistent. When you get really good at this, you can do it both ways with very little adjustment.
 

WBahn

Joined Mar 31, 2012
29,978
That a particular convention is not universally accepted says nothing about it's truth nor it's usefulness. If 999,999 cars on the road drive on the right side it is not an issue of the convention when some drunk drives on the left side.
I agree. Notice that I wasn't making any statement one way or the other about which view was "right", but merely commenting on your statement that people should "Always always use conventional current when discussing electronics with anyone else." The justification for this was because it was convention and that, "Convention: it's the convention! "

That argumemt is only valid to the degree that it really is the convention and, (unfortunately, in my opinion) it's not as widespread as those of us in the "conventional current" camp sometimes like to believe.

WTF am I talking about? Have you even used a digital logic chip?
Yep, I do seem to recall using a couple here and there. Come to think of it, I seem to recall designing a few here and there.

It has long been the convention in logic chips to denote current into the device as positive, and current out of the device as negative. Simple, clean, seems to make sense too, right?
Yep. And notice that the convention is independent of whether you are talking about an input pin, an output pin, a bidirectional pin, or a a power pin. This convention has nothing to do with "conventional" or "electron" current. It has to do with ensuring that automated tools, such as simulators, don't have to implement a bunch of complicated rules that, invariably, may still end up having problems because the input decks are ambiguous.

But what about the outputs? Check sheet 5, "recommended operating conditions." You will note that the device (a simple inverter) is spec'ed to source Ioh of -0.4mA, and sink Iol of 16 mA.
And????

This is completely consistent with the convention of denoting current into the device as positive and current out of the device as negative. When the output is HI, it may have to source a current OUT of the device of as much as 0.4mA; since currents OUT of the device are negative, this is -0.4mA. When it is outputing a LO, it may have to sink a current INTO the device of as much as 16mA; since currents into the device are positive, this is a +16mA.

The biggest influencing factor, for a very long time (pretty much since ICs were first being developed), on this convention are simulators. When you design a simulation block for a part you have no choice but to adopt the conventions of the simulator. Since most simulators neither know nor care whether an interface port is an input or an output port (and let's not even get into bidirectional ports!), you need a very simple convention and the convention that is used is usually very simple -- current INTO a port is positive and current OUT of a port is negative. They don't care if it is a source or a passive component.

So if you probe a resistor, you will discover that the current is positive at the pin that is at the higher voltage relative to the other. But you will also find that the current at the more positive pin of a voltage supply or a current supply that is delivering power to the circuit will be negative. Simulators neither know nor care whether something is supposed to be a source or a load -- everything is a passive device and current direction is defined according to the passive sign convention.

Now, many simulator interfaces, particularly GUI output services, have some stuff layered on them. For instance, the basic simulators usually only know about currents into specific pins. But some simulators allow you to ask for the current in a resistor or other part. This generally gets translated into the current into the first pin in the pin list.

Getting back to logic parts, one place to look for something somewhat "authoratative" would be the relevant JEDEC specifications document, such as JESD8-B. Unfortunately, it is hard to come across because JEDEC vigorously defends their copyrights and wants everyone to pay for them.

Interestingly, while they specifically state that all voltages are relative to ground except where noted, they make no statement at all regarding definitions of current polarities, which I think is a blatant oversight on their part. You are left to infer these from the tables themselves.

For the outputs it is pretty clear cut if you look at their test conditions. The V_OH test uses a current of I_OH=-100μA and the I_OL test uses a current of I_OL=+100μA. Since they definitely use conventional current (this is so universal in these kinds of documents that there is simply no need for that fact to even be stated) their current direction is that currents into the pin are positive. Their adoption for input currents is impossible to divine the specs table because they simply specify that the input current is to be less than ±5μA.

Now before you electron flow guys start jumping up and down, do note these are both spec'ed as MAX values. So one may be lead to believe the output MUST always drive that -0.4 mA OR MORE, since -10A is less then the MAX of -0.4mA, right?

Nope... somewhere in the general spec for TTL it's stated currents are evaluated using their magnitude only, so the sign disappears. (While checking a 74C04 data sheet it seems this absolute value convention has since been dropped.)
I don't see that any of this discussion says anything about conventional current versus electron current. The adoption of the IC manufacturers is that the polarity of the current at the pins, input or output, is such that conventional current flows into the pins.

Because the sign only has relevance for indicating polarity, the quantitative specs have pretty much always referred to the magnitude only.
 

hobbyist

Joined Aug 10, 2008
892
It's good to be familiar with both, but it seems to be easier to trace digital signals using conventional flow, since most is positive logic, and as far as analog circuits, conventional flow seems to overrule, because most all circuits are negative ground.
 

pstemari

Joined Sep 1, 2013
1
You'll drive yourself nuts with the "electron flow" convention once you get to anything involving electromagnetic activity: antenna theory, Maxwell's equations, etc. The right-hand rule becomes a left-hand rule, the textbook equations will all have sign flips on the current terms, etc.

For semiconductors you have both positive (holes) and negative (electrons) charge carriers, so there's no real reason to let that drive your thoughts.
 

MaxHeadRoom

Joined Jul 18, 2013
28,619
so I decided that a diode symbol looks like a speaker symbol. The bar represents the magnet and the wide end of the arrow emits electrons like an speaker emits sound.

.
Another revived thread!
I came up through the tube era, so I was taught electron flow pertaining to valves, then along comes the semi conductor Diode replacement for the tube rectifier and the darn check valve symbol was facing the wrong way!!
Max.
 

WBahn

Joined Mar 31, 2012
29,978
I learned that the diode symbol comes from a graphic depiction of a plate (the cathode) and a spray of electrons (the V) coming off of it from the center and spreading out before hitting the anode. It is just a convenient coincidence that this can be viewed as being an arrow pointing in the direction of current flow (where current is the flow of charge, not the flow of the charged particles that may or may not flow in the same direction depending on the polarity of the charge). Now, I don't know if this is the actual origin of the symbol or not, but it has always been an easy thing for me to remember. In fact, since I almost never have need to deal with the terms "anode" and "cathode" I haven't internalized them at all. So when I do need to consider them I think of the diode symbol and a "cathode ray tube" and that is enough for me to figure out which terminal in a circuit is the anode and which is the cathode.
 

Wendy

Joined Mar 24, 2008
23,415
This site uses electron flow. You have to learn both, so get used to it.

Necroposting is also frowned on, open a thread of your own and watch the fur fly!
 
Status
Not open for further replies.
Top