So looking back at your original post...
Some meanderings/observations/thoughts.....
Note - my fuel cap always seems to have a a slight positive pressure 0.5psi(?) when I go to fill up.
Anything over this and I guess the vent valve (76) opens to the canister - I'm presuming.

The canister would vent to atmosphere - eg: when you are stopped and filling your tank....about 1.5L/sec?
I'm not sure if AC (Activated carbon) extracts fuel - it must do...otherwise the vapours would be very strong - and dangerous.
Somehow this AC has to discard this buildup, I presume through fresh air into the AC when the engine is running and the -ve side of the manifold (pre SC) is open (purge control valve Y58/1).
The tank needs to suck air in also as you use fuel (max 0.05L/sec) or the tank cools down (possibly similar volume expansion/contraction rate of 0.05L/sec here). So the tank vent valve would have something like ....
+0.5psi = expell
-0.5psi = suck - it cant suck unless it is -ve (or it is pumped in - which it isnt).

The purge valve would have some sort of differential pressure setup...probably of a mechanical form - others may be able to add something here. eg: does it have wires attached?
when the manifold prs is -ve it opens and allows the canister to suck atmospheric*
when the manifold prs is 0 (stationary) the valve is closed and the canister would vent to atmosphere - eg: when filling the tank.
If its not an electrical setup - this would be the case all the time - if it has some sort of timer on it (electrical) then maybe it would other ECU ties.

I've been meaning to put some more thought into this, but have been a little busy with a home made OBDII logger setup I'm working on. So just some thoughts put out there for discussion.

* would this add to our "going lean" issues if we are pulling above stock cfms and our vacuum ex throttle is greater? - I dont think it would be a significant volume compared to 600cfm.