[RLCrisp]

Looking at the Class A biasing method for PIMETA, I have a question. I understand the point of the current source (to force the opamp's output stage single ended using the faster BJT's) but am a little confused on the effect that the cascode has on the load resistance of the opamp.

The BUF634s have imput impedances in the megaohms range, which is good from a loading standpoint on the input stage (in my case OPA132), but doesn't adding the JFET cascode drastically reduce the load OPA132 would see?

If you try to calculate input resistance it seems to be (Vo-Vneg)/I = 4k (assume a +/- 12V source with no output voltage and 3mA current source for 2N5484 from Art of Electronics Fig 3.17) from the opamp output to the negative rail.

This is still a pretty good load resistance but reading Tangent's Class A Biasing page: "The other benefit of using a JFET instead of a plain resistor is that the input impedance of a JFET is up in the megohms range, so the op-amp still sees a near-ideal load situation." This is true looking into the gate - but that doesn't have anything to do with the loading resistance of the cascode in this application, does it?

Thanks for any help - sorry this was so long.

 

[RLCrisp]

Found the answer in an app note I think. Talks a lot about the small signal output conductance and how to calculate it. Then taking output resistance as the inverse of output conductance got a value of almost 2.5 megohms for a JFET cascode of 2x 2N5484.

The app note gave output conductance with JFET cascode and no source resistor as:

go=(goss)/(2+gfs/goss)

The app note is at: http://www.vishay.com/docs/70596/70596.pdf

Now I can feel relatively confident about putting it in the pimeta =)