[steinchen]

atm I'm using hand matched 2n5484 on Q1 and Q2 on my PPA, R10=1k and the pot is 2k. With the pot turned all the way down I get a maximum bias of 1.7mA +- 0.1mA on the channels.

I'd like to try a higher bias, what do I have to alter at the cascode ? (picking jfets with higher IDSS I assume) Which of the two jfets determins the max cascode current ? Or is it the difference of the two IDSS values ?

 

[stadams]

Quote:

Originally Posted by steinchen atm I'm using hand matched 2n5484 on Q1 and Q2 on my PPA, R10=1k and the pot is 2k. With the pot turned all the way down I get a maximum bias of 1.7mA +- 0.1mA on the channels. I'd like to try a higher bias, what do I have to alter at the cascode ? (picking jfets with higher IDSS I assume) Which of the two jfets determins the max cascode current ? Or is it the difference of the two IDSS values ?

I went back and read my previous post and realized I was in error. If you are using the 2N5484 you could very well be maxed out for your bias. If you can find it, try the 2N5486 or any other JFET that has a higher Idss than the 2N5484. Sorry for the error.

Later,

 

[tangent]

Quote:

what do I have to alter at the cascode ?

Select Q2 for higher bias at minimum, and you may also have to increase the IDSS of Q1. I've used PN4392 for Q2 before, for instance, when I want a wide adjustment range. I seem to recall that I got a 0.7mA to 7mA range that time.

Quote:

Which of the two jfets determins the max cascode current ?

Both do, together. Read this, in particular where it describes how the two JFETs interact. If you want more detail than that, I think you'd have to pull out your calculator and find some JFET equations. Or, fire up SPICE...

 

[stadams]

steinchen,

If you haven't looked at these application notes already, check them out. This one in particular may shed some light on your situation.

The information that tangent has on his web site is very informative also. It sure is nice to have people around here that are willing to shed some light and share a little knowledge.

I see that tangent also has a link at the bottom of op-amp bias page to the same document from Vishay/Siliconix.

Later,