Efficiency Puzzle: Wood-Powered Tree-Farm?

[Limdood]

the problem is that your system isn't closed.  Trees gain energy from soil and water as well as light, neither of which is simulated in Rimworld.  That means as a simulation, you can't really have a completely "closed circuit" because it requires outside input, such as soil (since soil, rich soil, and gravel would all give vastly different outcomes to the experiment).

[Thyme]

Soil and water provide nutritients, but no energy. That comes solely from the sun.

[Hans Lemurson]

Well, my tribe just starved to death on an ice sheet waiting for potatoes to grow.  I underestimated how cold rooms get and how much the heat is diffused when you make them large.

I think I need to do some serious research into how Heat works in this game.
...
Ok, some important observations:
-Holes in the roof drain a LOT of heat!  Crazy, huh.
-Insulation helps some, but holes have the biggest impact.  Double-walling buys you a couple extra holes.
-Double-walling increases the temperature differential of a given heat source by 50%.  This is comparable to doubling the heat sources.
-Triple-walling gives no benefit over double-walling
-Campfires create a max room temperature of 28C, torches at 23C
-Campfires burn 10 wood per day, torches burn 2.2 wood/day
-It takes 6 torches to equal the heat output of 1 campfire (campfires win)
-1 Fueled Generator powering 5 Heaters generates over 2x as much heat as 2 campfires, for roughly the same wood/day.  Electricity FTW!!!

Skylit wood-heated arctic tree farms seem do-able, but I don't think campfires are up to the task.

One major challenge with skylight farms is that the temperature is near its lowest when the daylight begins and the plants wake up, so you stand to miss out on a lot of growing-time if your heat source can't handle the lows.

[Thyme]

I just got my first harvest of tundra grown wood! Picture below. My attempt at heating only during growing hours didn't work, my trees disappeared. A fellow player in Phi suggested as cause that a some of the trees die when frost hits. Doing so every night is the road to financial ruin ...
So, I resorted to permanent heating and connected the sunlamp via a time-of-day switch to the main grid. The solar panels were added to take some of the load from the sunlamps (got more greenhouses). Powering the sunlamp directly with solar panels would work, but is suboptimal. Solar panels have 900W at 0600 (start later) and are close to 0W at 1900 (end way too early). Batteries would have helped, but I like to avoid Zzzt!.

Using a solar panel to cover most of the sunlamps drain, another power source combined with a battery to sustain the heating is definitely more resource friendly and efficient than Hans' approach with fueled generators. Using the sunlamp 24/7 is not recommended, the reasons are obvious. A switch shouldn't be neccessary for that if you make two independent circuits.

Other notable things to mention:
+I looked for an optimal spot with Zhentars sunlamp designator. I think that's a part of his vanilla tweaks. It sums the fertility of all tiles that would be covered by a sunlamp if you where building one where your mouse is. I looked for the maximum number of trees afterwards. Great tool for Ice Sheets!
+If you like it more vanilla friendly, Zhentar has a switch that assigns itself automatically for flicking in the morning/evening.
+The third heater is currently a backup. Climate cycle was currently at peak temperatures with -10°C summer, but will go to -50/70°C summer/winter soon. Might need a fourth.
+I placed 18 growing zones to work around possibly stupid growers ;)

[attachment deleted by admin due to age]

[Hans Lemurson]

Using a solar panel to cover most of the sunlamps drain, another power source combined with a battery to sustain the heating is definitely more resource friendly and efficient than Hans' approach with fueled generators. Using the sunlamp 24/7 is not recommended, the reasons are obvious. A switch shouldn't be neccessary for that if you make two independent circuits.

I think you're missing the point.  My question was not "What's the best way to grow wood in woodless lands?", it was "Is it theoretically possible to extract net energy from a closed system?"  And the answer turned out to be yes.

Only a total fool would try to power their Sun-Lamps with the very wood they are growing. 

[Thyme]

I answered that question in post #7 already, rice in hydros is much more efficient than wood. Corn on rich soil should top even that.