I was looking at some of the bipolar devices in the PDK again and noticed that there is a statement in the PNP section that says: "No deep n-well exists in this device; the collector is the substrate." But the drawing shows a P-well, as if the device could very well be placed inside a Deep N-well. Would it be possible to place the device in a deep N-well? I cannot see why not, but i could be wrong.

@Christoph Weiser Inside a deep nwell, all the nwells connect directly to the deep nwell. Since there is no space between the nwell and the deep nwell, the pnp transistor would be p-diffusion/nwell+deep-nwell/psubstrate. You might also be able to do a p-tap+pwell/deep-nwell/p-substrate PNP bipolar transistor (maybe more of a lateral transistor), but I don’t know that anyone has ever characterized such a critter.

That makes sense. Thanks for the quick response.

The nwells inside a deep-nwell are not isolated ?

No they always connect vertically as pointed out by David.

I never realized that. I thought I could always place a deep n-well under any IP core to isolate it from noise and such, but I never realized this wasn't "transparent" and that this would connect all the n-wells inside of that IP ...