[R-390] Cosmos PTO, spring-loaded linearizing core?

[Tim Shoppa]

Jim recommended:
> Fix the wife some chicken soup, clean up the kitchen and get the kids
> under control first.

OK, made that much progress :-).

But now I have more time to write my real question:

How the heck is the linearization screws/plate/core/inductor supposed
to work in real life?

"In principle" it's a spring loaded thing such that the core follows
the 40-some linearizing screws through the rotation.

But this is what I find in my Cosmos:

The linearizing inductor has a core in it.
The core is on a leadscrew.
The leadscrew goes through a threaded hole in the metal base of the
  inductor.
On the other side of the leadscrew is a plastic nub that, I presume,
  is supposed to follow the 40-some screws.

What I don't see:

Any way for the nub to move the core in and out of the little linearizing
  inductor.

There's no spring. Nothing slides.

Maybe, just maybe, the nub compresses/flexes the metal base such that
the tiny corrections are made. But it doesn't seem likely.

Maybe, just maybe, the pressurized nitrogen in the PTO provides the
restoring force and the sliding is just the screw wiggling in its
threads. But that doesn't seem likely either.

Obviously my knowledge of how the linearizing is supposed to work has
a huge gap in it, because I don't see how the screws move the nub
or how the nub moves the little core.

Can anyone PLEASE correct my mistaken reasoning here?

Tim.