Thermodynamically impossible without an input of energy. FWIW.
Thermodynamically impossible without an input of energy. FWIW.
Ventillation to exhaust waste heat is a major concern.
From discussions of net-zero housing in Alaska (Thorstein Chlupp), my understanding is that water tends to flow from hot to cold -- it will condense along a cold surface. A condenser + sump system where some engineering mechanism keeps the inner tunnel wall from remaining at equilibrium temperature with the interior, might work, though how cold that would need to be to keep humidity manageable would be a question. Routing the cold-side air ducts such that they waste heat to the subsurface and draw heat from the tunnel wall might be one approach. Further cooling through water (fresh, sea) is another possibility, though both present significant issues, scarcity of fresh in Los Angeles, and general nastiness of sea,
(The only time this may be untrue is when dealing with very small things like remote sensors, where running a wire might actually be more expensive than harvesting the microwatts needed locally. But this is a fraction of a fraction of a percent of total system needs.)
In some ways, this is the inverse of an evaporative cooler, where the incoming air is saturated with moisture to lower its temperature before being vented to a warm space to be cooled ("swamp coolers" are best suited to dry climates, and have humidity issues). Here, instead, cool (possibly chilled) air could be ducted through the warm space, with the intent of condensing moisture onto the ductwork (for drainage by other means) prior to being vented into the space.
You're trading cooling capacity (condensing water warms the cold air supply) for humidity control.
I'm not trying to imply that this is practical, only that a latent energy flux might exist.
Check valves absolutely do consume energy. Something has to resist the spring forcing them closed, or they would never open. They are convenient in that their energy is sourced from the very thing they are controlling, but they are energy-consuming devices.
There are easier ways to get moisture out of a tunnel though.