Edit, for some reason when I try to post the just image I’m told it can’t be found, so you’ll have to click the link to see it full size.
In this image, why are the IBE and IBC currents shown going against the diodes?
Positive current direction in a diagram is arbitrary. If the defined direction goes the “wrong way”, the assigned current will just take on negative values. So I assume it’s to make the associated equations look nicer. Or whoever made the diagram thought it was tidier to have all currents go into the central node.
It is also possible to get a current flow opposite the direction of a diode; you’ll just get a significant voltage drop going that way. It’s been a long time since I’ve studied semiconductors, and then not in much depth, but I think that might be important for a model of a transistor in terms of diodes.
I was under the impression that Ie=Ib+Ic. Ie should be flowing the opposite direction.
The direction is arbitrary. You could just as easily write it as Ib+Ic+Ie=0 which conveys the same thing but has Ie in the opposite direction.
It’s similar to Kirchoff’s current law, which you can state as the sum of all currents entering a node must be equal to the sum of all currents exiting the node, or you can point all currents into the node and say that the sum of all currents must be equal to zero (which then obviously implies that if some currents are positive then others are negative, i.e. exiting the node).
The latter version is simpler to express mathematically: \sum_{k=1}^n I_k=0
Since the Gummel-Poon model was designed for computer modeling, I suspect that they defined their currents in the direction that made the programming / modeling easier.
(ninja’d a bit by @Crafter_Man while I was trying to get my math symbols right, but oh well, I’m posting this anway)
An ideal diode with a “reverse” voltage across it will result in zero current (a.k.a. open circuit or infinite impedance). When you do that to a real diode, a very small amount of current will flow through the diode, usually somewhere between 10’s of nanoamps and a few microamps. It depends on the magnitude of the reverse voltage, the type of diode, and the temperature.
An ideal conventional diode, maybe. But there is also the Zener diode, which is specifically designed to allow reverse current at a specific voltage (they can be used as a voltage reference, among other things). An ideal Zener diode would have no resistance or voltage drop in the forward direction, while in the reverse direction would have infinite resistance up until the threshold voltage, after which it has zero resistance.
Very true. I should have mentioned that I was referring to regular ol’ diodes in my post above.
All diodes are Zeners, some are just better at it than others. You can abuse any semiconductor junction in ways that the usual models don’t like. Sometimes with useful effect.
Some colleagues of mine many many years ago claimed to have discovered the NED. A noise emitting diode. Unfortunately it is a one shot device. Very effective at drawing attention to errors in design and wiring.
That’s almost as good as when one of my students discovered that Magic Smoke is, apparently, hallucinogenic. He’d just accidentally let out the magic smoke from some capacitors, or so he told me, when lizards started crawling out of the wall sockets.
I think I got it. Thank you everyone.
Electrolytics contain some pretty weird organic stuff. But that is just a little disturbing.
And why is it always lizards?
BJT - bipolar junction . That means three layers, NPN or PNP semiconductor layers.
A BJT is not made out of two diodes. It is made out of three layers of semiconductors, but two diodes would be four layers… Its just drawn that way, for consideration of whats going to happen if used out of normal , eg you don’t connect the third terminal, or switch the voltages around.
What actually happens is that the current through one junction in the transistor creates carriers in the reverse biased junction, because the of careful design of the layers of the semiconductor … the shared semiconductor , the base, is very thin and sandwiched between the two other semiconductors.
With an NPN - electrons are pushed into the reverse biased junction.
WIth PNP, its holes.
You could make a LED Transistor from two diodes. eg this is effectively the IR transmitter, IR detector paired devices… the IR transmitter is a LED … a diode that emits IR in forward conduction … the IR detector is a reverse biased diode sensitive to the IR , with the IR transparent holes giving good access, as in maximal surface area, of the semiconductors. The IR then creates conductivity in the reverse biased diode, and hence current in the reverse biased But this would be creating electron hole pairs in the receiver diode, not limited to one or the other.