some JFET questions

[tempus]

Hey all;

A couple of quick questions regarding JFET preamps:

1. From what I've been reading, it appears that the output impedance of a JFET preamp is governed primarily by the internal output resistance of the device. Is this correct?
2. Is it correct that the internal output resistance of the device is given by the inverse of the output admittance of the device i.e., 1/Y_os?
3. At least a few of the datasheets I've been looking at do not have a figure listed for output admittance, but using the same unit (umhos) they list a forward transfer admittance or transconductance. Are these the same thing? Also, is mmhos the same thing as umhos?

Thanks

[antonis]

1. Correct but for infinite Drain resistor.. For real world amp circuit is dominated by the much lower value of the later..
R_out = R_Drain//r_o, where r_o= V_A/I_DQ with V_A been the "analogous" of Early voltage for BJTs.
(V_DS - I_D slope = 1/r_o.. you can search for "channel length modulation" for more info..)
2. Correct in the mean of admitance is the reciprocal of impedance..
3. Transconductance is a transfer coefficient relating output current to input voltage and can be thought as representing the gain of the transistor..
g_m=2 X square root(K_n x I_DQ), where K_n is conduction parameter (a function of the channel width-to-lenght ratio)..

[PRR]

What he ^ said.

Yos matters above 100kHz where we can use tuned circuits and get very high load impedance.

In Audio we use *resistors* (or rarely, fat iron) to bring DC into the FET. We can never get near typical Yos.

A typical spec will be 10uMho at about 5mA. 10uMho is 1/100k resistance. Say we pick a similar value drain resistor, 100k. 5mA in that says a 515V power supply (add a little for the FET). This is inconvenient. Also who wants a 50k output impedance?? We might end up nearer 2k load (10V 5mA). Then Zout is 100k||2k = 2k within 2%. Even more often we would run a Voltage Amplifier nearer 1mA because Gm/Id improves as current is lowered.

[antonis]

Also, is mmhos the same thing as umhos?

Of course, not..
(mhos is the plural for mho which is the converse word for ohm..)
So, m(mhos) = μ(mhos) x 10³

P.S.
Pay attention on Paul's symbol discrimination..
(M used as unit inceptive capital letter for discrimination from unit multiplier m (milli)

[ElectricDruid]

It's awkward. Alternatively we could use "S" for siemens, since mhos isn't an official unit and hasn't been in years.

   uS = microsiemens =uMhos, mS = millisiemens = mMhos

but then you can confuse things with seconds:

   us = microseconds, ms= milliseconds

Perhaps context helps for that, though.

Ultimately, I like the upside-down omega best of all. Unlikely to mix that up with anything else, and it provides a nice visual reminder of what it is we're talking about.

   uƱ = microsiemens =uMhos, mƱ = millisiemens = mMhos

[antonis]

Hmmm...

Tom, if you're able to use "Ω" (inverted or not) then you SHOULD be able to also use "μ" for micro..!!!
(and not that $#@@*(&^ "u" which points to something like "ultra" or so..)

[ElectricDruid]

True. I was being lazy. I copied the upside-down omega from a google search, but I didn't bother for the "μ".

Can you forgive me?

[antonis]

Only if OP forgive us both for spamming..