Quote: "Not quite! There are no such thing as PNP or NPN MOSFETs; those two types only apply to bipolar transistors. And indeed, CMOS circuits are only constructed from MOSFETs, not bipolar transistors."
Ah, I was just using the terms interchangeably since they effectively mean the same thing (and I didn't know NPN/PNP didn't apply to MOSFETs). NPN/N-channel blocks current flow with 0V at the gate, PNP/P-channel allows current with VCC at the gate.
Quote: "I actually feel for you quite a bit, since that is not a fun feeling. I remember days staying up incredibly late, just trying to come up with a complete understanding of a particular subject. Walking away from the Internet with an incomplete understanding of something is a little unnerving, at least for me."
The unnerving part is that I've spent hours and opened 100s of Chrome tabs and STILL can't find the information I'm looking for. It feels like everyone else knows something I don't, and they don't understand how I could not know it.
Quote: "If I recall correctly, this is one of the silly things about electronics nomenclature. Electrons actually flow from negative to positive."
OOOOOOOOHHHH
DUH!! *facedesk* I actually KNEW that, but I needed someone to remind me of it, thanks! It's a very silly naming system. In general speech, "negative" means a lower number or less of something, whilst "positive" means a greater number or more of something. Naturally, you would expect the positive side of a battery to have more electrons. Unfortunately, the positive side of the battery isn't called positive because it has more electrons, it's called positive because, physically, electrons are negative. A build-up of electrons is a negative charge, and therefore the side where the electrons collect is called the negative side and the other side is called positive. Am I remembering my Science textbooks correctly?
Quote: "Hehe, you are correct, ALL logic gates do indeed have to be connected to VCC and GND. You just don't see it on logic gate symbols because, well, it's a little redundant. The connections are there on the actual circuit, but that's taken for granted, so it keeps things more neat to simply omit the connections in the diagrams."
Really? Hhhhh... this is the kind of thing that's not explained to beginners, yet can cause great confusion and misunderstanding.
Quote: "Here is something very important to understand about CMOS logic gates: within a CMOS logic gate, you literally have two voltages, those being VCC and GND. There is pretty much, for practical purposes, no in between."
Are you sure? Because from my experiments and some further research, it seems to me that the Gate voltage (whoops, nearly typed current there!) has to be higher than the Source voltage to full turn on the MOSFET. In my Proteus simulations, I'm using an
IRLML2502 N-channel MOSFET for the simplest test I can come up with: switching on the transistor to power an LED.
In the following image, the circuit is behaving as expected. The switch has 5v on one side but is open and doesn't allow the voltage to the Gate on the MOSFET. Well, mostly as expected; it's allowing a tiny voltage through (0.4v), even though the Gate is not powered, but for my purposes I don't believe that matters at all.
Now I've closed the switch. According to the datasheet for the MOSFET (linked above), the Gate Threshold Voltage is between 0.6v and 1.2v when the Vds and Vgs are equal and a load of 250uA. Given this, I would expect to see the Gate be fully open with a voltage of 5v for both Vds and Vgs. Yet, as you can see in the image below, I'm only getting a partially-opened MOSFET, which I assume would be getting pretty hot right now if this were done in real life.
If I increase the supply voltage to 6.2v (1.2 above 5v), I see the expected results. I was going to rant on a with a longer theory here, but something has just occurred to me. Here's the picture anyway:
Ok, so you know what occurred to me? I just realised that I was typing
Vds and
Vgs and claiming specific voltages, but I'd never actually measured it. I'm assuming that the voltage "probes" (the blue arrows) are assumed to be connected to ground of the other end (because voltage is a measurement of how strongly electrons want to go from one point to another, right?)... which means I'm only measuring my voltages to ground. I'm never actually measuring voltage from Drain to Source or from Gate to Source. So I went searching, and quickly found a voltmeter tool. I stuck a couple of these on and found that the Vds was only 1.12V, as seen here:
The issue now is that I'm not sure why there's such a low voltage for Vds and Vgs. It needs to be at least 1.2 to fully open the MOSFET. Also, I stuck the VCC at 3.3v to test, and the Vds/Vgs barely changed at all (I think it went down to 1.05v or something). As far as I can tell, the only reason the MOSFET fully opens with a VCC of 6.2v is because it just happens to bring the Vds/Vgs up to almost 1.2v.
What am I misunderstanding here that's meaning I'm not getting enough voltage on Vds and Vgs to fully open the MOSFET?
Sidenote: Before I put the wire with resistor R10 in there, the LED refused to turn off. Putting that wire there was a near-complete guess, but it was based on something I thought I'd read a few hours ago that mentioned capacitance. In an attempt to apply logical reasoning using as accurate an understanding of how electricity works as I could, I determined that when the switch was open, the Gate was probably becoming charged, either due to simulation inaccuracies or due to random static charges (and the fact that if I closed and then opened the switch, there was nowhere for the electricity left in the Gate to go), and was keeping the Gate open even when the switch was open. So I stuck a wire and a resistor in there and it worked.
Quote: "Oh yes, some sort of scattered side advice before I go further: for now, I would recommend forgetting about drain and source, until you have a better understand of electronics. For now, consider a MOSFET to have one gate terminal, and two coequal terminals that current may or may not flow between, depending on the voltage at the gate."
Well that may be a good idea except that transistors are polarised
You've gotta have differentiation between the two pins! But don't worry, I've learned
why you connect the Drain to the positive side. Like with maths, once I've learned the
why, the concept cements itself pretty quickly in my brain
Quote: "I remember this by noting that the P channel has a little circle on the gate, just like the letter P."
Now
that's something I already have a firm grasp of. I already read that the little circle means "inverted" and that bit of info managed to stick
Quote: "I hope this is clear as possible, and hasn't confused you more."
It definitely hasn't confused me any more! On the contrary, you've done a brilliant job of helping clear my mind and sort out the mess of information I've tried to stuff in over the past few days
The way I see it, over the past few days I've been doing what I call "forced learning". It's what I do when I want to learn something, but don't want to do any kind of courses to learn it. I just begin researching anything and everything on the subject and just absorbing information until I have just enough of the base parts to begin
using the information. Then, as I attempt to use it and find I have gaps in my knowledge, I research more specific parts and forcibly memorise the info. This method has served me extremely well for learning new programming languages in the past, but it has its disadvantages.
The biggest problem with learning like that is you learn everything in a very unstructured way. When the subject is relatively small (like the syntax of a new language), I still end up with a messy cloud of random, unstructured information in my brain, but I'm able to use it and it settles down into a rational structure pretty quickly. When it's a big subject like electrical engineering, however, that cloud becomes a huge, thick, black disaster of information storage. My brain just says, "OOH no, there's no way I can sort that lot out, you're on your own!", and promptly shuts down. This is when I become confused.
The only way to unconfuse myself is to have someone explain it to me by answering specific questions. I take the most easily-grabbed piece of info from the cloud and ask you about it. You answer, and my brain can then sort it out. The more pieces are sorted and placed in a structured way in my brain, the more my brain can sort out the rest of the info itself. It just needs those initial "hooks" to hang the rest of the information on
I'm still a little uncertain that those "hooks" are actually correct, which is why I'm still asking you questions for the moment
I greatly appreciate your very clear explanations. Thank you!
Now to go and attempt to rationalise exactly what those two voltmeters are actually measuring and why they are behaving as they do...
Wait, no, it's twenty to one in the morning, time for bed!
