EEVblog Electronics Community Forum
Electronics => Beginners => Topic started by: c4757p on October 17, 2012, 02:40:25 pm
-
First of all, hello! This is my first post here.
http://socrates.berkeley.edu/~phylabs/bsc/PDFFiles/2N3819.pdf (http://socrates.berkeley.edu/~phylabs/bsc/PDFFiles/2N3819.pdf)
I was wondering if somebody here could help explain this to me. In this datasheet, there are two graphs: "Common-Source Forward Transconductance vs. Drain Current" and "Output Conductance vs. Drain Current". Both of these graphs are labelled as measuring gfs against ID, under the same conditions (VDS = 10V, VGS(off) = -3V). They are clearly very different plots. I've been scouring the Internet and cannot seem to find anything that explains the difference between the two gfs's - all the material I can find on "output conductance" seems to be describing the first graph.
-
Sorry for the bump, but... anyone?
-
Some stuff to check out:
Conductance:
simple definition = The inverse of resistance
http://en.wikipedia.org/wiki/Electrical_conductance (http://en.wikipedia.org/wiki/Electrical_conductance)
Transconductance:
simple definition = The ratio of the current change at the output port to the voltage change at the input port.
http://en.wikipedia.org/wiki/Transconductance (http://en.wikipedia.org/wiki/Transconductance)
This site has examples of calculating both with respect to fets:
http://ecee.colorado.edu/~bart/book/book/chapter7/ch7_3.htm (http://ecee.colorado.edu/~bart/book/book/chapter7/ch7_3.htm)
Siliconix is Vishay these days, but that part is made by a by a bunch of companies. I've always had good experiences calling up tech support and asking questions about data sheet stuff. It's usually not that hard to get a hold of someone who knows what they are talking about. They don't care if you are just some dude or a huge company. This wouldn't be the stupidest question they had to answer today either.
-
OK, if no one else will.
The common source forward transconductance is the two-port parameter y21 in, as it says, common source configuration. Thus the definition is
y21 = i2/v1 i.e. in this case, drain current vs gate-source voltage.
The output conductance is
gd = dID/dVDS
i.e. change of drain current vs drain-source voltage.
-
Thanks for the links. I do know what conductance and transconductance mean (and I've been through the Wiki pages just in case), but I'm having a bit of trouble understanding how they are related in this case (most confusing is that both plots claim to be measuring gfs). Perhaps the third link will help. As I understand it, "output conductance" would mean something like iD/vDS, but if I measure that I get nothing like the plot in the datasheet.
I didn't realize tech support at these companies would answer this kind of question, thanks! Maybe I'll give them a call. I don't even really need an answer specific to that part - that was just a random datasheet I picked that had the plots in it. I really wanted to know more for the sake of characterizing a part that lacks those plots - and just for the sake of understanding. I suppose anybody who makes JFETs could answer.
-
Kremmen, thanks! I'm going to rig up a quick test to make sure I'm not misunderstanding the explanation, but that seems right.
-
Wait - if the output conductance is related to VDS (as I thought it was), then how can they have "VDS = 10V" marked on the plot? Wouldn't it have to be variable?
-
Ya, sorry about the basic stuff. You never can tell where someone is coming from here. Plus a lot of people that read this post might not know what they meant before.
As far as reaching someone technical from the manufacturers, I know Vishay is pretty good actually. I had a multi-day correspondence with a Vishay/Siliconix engineer about a latch-up problem I was having with an analog switch. They even set up the chip in their lab and duplicated my conditions. At least that's what they told me :)
-
Ya, sorry about the basic stuff. You never can tell where someone is coming from here.
I can't even always tell where I'm coming from! I'm mostly self taught, so sadly I've got a fairly thorough smattering of some pretty advanced stuff, with some huge basic holes in between. I'm studying EE in school, but they just focus on the boring digital stuff. I've always loved analog.
That's pretty incredible about Vishay. I'm going to have to give them a call. Might be a bit late today, since it's already 4pm here and I'm tied up until 7:30, but I'll try tomorrow. Guess I'll keep banging on my LTSpice analysis setup to see if I can get a plot resembling theirs in the meantime.