For quite a while, I memorized the polarity of MOSFETs by thinking "Ok, when open, current flows into the drain of the device, and comes out of the source. Logical."
Now I finally used a p-channel MOSFET and gladly found out early enough that I a wrong with this in general. So what is the actual story behind those names?
I also noticed a similar misconseption with NPN and PNP BJTs ...
I can remember that the IBM part number for a SN7400 is 2392102 but I can never remember MOSFETS.
I remember where I put the data sheet, though. On my Dropbox...
terryking228:
I remember where I put the data sheet, though. On my Dropbox...
Well, I put it in my Zotero, but unless it contains a schematic with an example, it does not help me with the question where to put which terminal. I also don't think I will have trouble MEMORIZING it in the future (though that was admittedly my subject ...). I know that V_GS is an important quantity for both p and n-channel, and that n-channel can easily do low side switching, so I know that source goes to the lower voltage for n-channel and to to the higher voltage for p-channel. That is actually how I noticed in the first place. I wanted to find out "V_GD(th)" for the p-channel MOSFET and only found V_GS(th).
What I really wanted to know is what the hell is going on there. Why the name? Doing a little more research, I finally found out myself that it is about the doped zones, where "source" is the source of the charge carriers that drift into the "drain" zone.
I think I will go for the V_GS thing as a mnemonic from now on 
In a P-channel MOSFET, the channel (the region where the current flows between source and drain) is formed by p-type silicon, i.e. the charge carriers are positively charged holes.
The channel of an N-channel MOSFET is made up of n-type silicon, so the charge carriers are negatively charged electrons.
The 'source' is the source of the charge carriers, and the drain is where they flow to. So it makes sense that in most circuits, the source of an N-channel MOSFET is at a lower potential than the drain, and vice versa for a P-channel MOSFET.
When a charge is applied to the gate, that is opposite to the charge of the charge carriers of the drain and source regions, these carriers will be attracted towards the gate. The charges in the region next to the gate, that have the same charge as the applied charge to the gate will be repelled.
This causes a channel to form.
In other words, in a P-channel MOSFET, you need a negative voltage on the gate, to attract the positive charge carriers, and a positive voltage on the gate of an N-channel MOSFET.
Pieter
1 Like
PieterP:
In other words, in a P-channel MOSFET, you need a negative voltage on the gate, to attract the positive charge carriers, and a positive voltage on the gate of an N-channel MOSFET.
That's pretty much how I remember it. Vgs needs to be negative to turn on a p-channel, so source is connected to power and the gate is driven low (negative compared to source) to turn it on. N-channel is the opposite, so source is connected to ground and the gate is driven high to turn it on.
Small logic MOSFETs are typically completely symmetrical - the source and drain are interchangable
and there is a separate substrate connection it you want it. As in that diagram above.
Power MOSFET's are very very different, the source and drain are completely different, and all
the action happens between source and gate. There is no substrate connection because the
drain is the substrate, and current flows through the chip, not across. The power MOSFET also
has a diode between drain and source which is a necessary part of the sandwich construction.
An n-channel MOSFET has an n-type channel, which means the gate has to be positive w.r.t. the source
to attract electrons and form the (n-type) channel. All charges reverse in sign when you swap between
n-channel device and p-channel device (just as they do when switching between NPN and PNP).
I can remember SOT-23's... I've dealt with so many of them. And they more or less followed the BJT pattern for them with FETs - G=B S=E D=C. I can just look down at the part and "see it".
Edit - Oops that wasn't the question... ok. I guess I'm no longer fooled by names. I basically ignore them. They are just names to me.
In normal usage, the charge carriers will flow from the source to the drain. There are a few circumstances where this is not true (like reverse polarity protection or synchronous rectification), but it holds in the vast majority of cases.
In N-channel devices, the charge carrier is free electrons, in other words a negatively charged particle. Negative charges flowing from source to drain is indistinguishable from positive charges flowing from drain to source.
P-channel devices are a little easier, because the charge carriers is holes, which have a positive charge just like conventional current. In normal orientation, the positive charge flows from source to drain.
In both cases, the naming convention is consistent based on the behavior of the primary charge carrier used by that type of MOSFET. The apparent backwardsness comes from the fact that the charge carriers have different polarity in N-channel and P-channel MOSFETs.