cap mult (again) at low current.

Any enhancement mode MOSFET with highly-enough doped ohmic contacts is a good candidate to try, but you are not going to like the low-frequency noise. The first one I used years ago was the 2N7000. Si-based parts seem to tend to develop a funny threshold at low Vds, origin of which I've been too lazy to figure out.

GaAs MESFETs and HEMTs are equally likely to work, including the NE3508 which does. Ditto for their noise.

Regards, Mikko

Mikko, since you're here, there's a point I've been meaning to ask about.

Carriers freeze out in CMOS parts as well.

I sort of understand what happens in BJTs--the base and emitter become insulating, so you can't pull carriers out of the emitter, since that happens at the edge of the depletion region, and there isn't one at low temperature, because the chemical potential goes to about zero when the dopants recombine.

In CMOS, you're in more of a photoconductor type situation, where the carrier emission occurs at the contacts, so as long as the Fermi level is in the right place, it still works even at T -> 0. Right?

But it seems as though you have to dope the contact regions heavily enough to delocalize the impurity electrons even when they aren't ionized, because otherwise you don't have a conduction mechanism. How do the carriers get out of the contact region? They have to get into the normal conduction band somehow. Or do they get emitted directly from the metal into the conduction band, with the high doping just serving to get rid of the Schottky barrier?

Cheers

Phil Hobbs

Something like that, I think. The doping concentration has to yield a degenerate state, so it doesn't freeze out. The Schottky junction also has to be non-rectifying, but as it happens, aluminum makes fantastic ohmic (and self-doping, if you like) contacts, so that's easy.

I wouldn't mind a refresher in the subject, myself...

Tim