LTP with CCS and feedback

[dgta]

A sometimes-used topology involves a LTP with CCS in the tail, but with the addition of feedback from both anodes to the CCS. The earliest example I saw is in Curcio's '89 article in GA for his cascode LTP.

The idea is to further balance the LTP by taking signals from both plates through a summing junction and applying it to modulate the CCS.

Curcio's CCS was a simple 1-transistor affair and presumably not real stiff. That may explain the usefulness of this feedback scheme. However, a modern, very stiff CCS doesn't need any help to balance the LTP. So, would you still use this feedback feature? What for? Increased CMR?

Anyone currently using it, or designing with it, can you comment?

[SY]

Yes, CMR is the big deal. You can improve it in Curcio's circuit by running a battery (or string of batteries, you want 80V or so) between the cathodes of the bottom tubes and the grids of the top tubes. I changed out the bottom tubes for FETs (2N5566), then used a 9V battery between cathodes and grids.

[revintage]

Is there really any AC-feedback? The AC at each anode should countertact each other in the junction between the two resistors going to the PNPs base.

[kevinkr]

Originally Posted by revintage

Is there really any AC-feedback? The AC at each anode should countertact each other in the junction between the two resistors going to the PNPs base.

That's the whole point, the connection to the CCS assures that there isn't.. (forces the AC to zero at the summing point - it goes with saying that those resistors should be well matched.)

[SY]

The point of that feedback is increased common mode rejection (and range).