Resources and ISRU

[gotmachine]

The Kerbalism paradigm regarding resources and ISRU is :

• LiquidFuel is an abstraction of methane (CH4), but also an abstraction of Kerosene/RP1 (C10H22)
  • Non-cryogenic
  • Medium ISP, high density
  • Used for rocket and jet engines
• Oxidizer is an abstraction of hydrogen peroxide (H2O2)
  • Non-cryogenic
  • Functionally equivalent as LOX as far as engines are concerned
• MonoPropellant is an abstraction of Hydrazine (N2H4)
  • Non-cryogenic
  • Low ISP, high density
• Additionally, I should mention LiquidHydrogen (H2)
  • Cryogenic
  • High ISP, low density
  • Relatively widely adopted in the KSP modding ecosystem

This works relatively well, allowing to somewhat fit real-world chemical processes into the KSP abstractions, and to provide a somewhat realistic ISRU scheme. This being said, there are issues :

• Carbon is quite difficult to obtain. Since we consider Ore to be generic regolith, we assume it has a very low carbon concentration. Atmospheric CO2 (Duna...) is also a very inefficient source. This has the consequence that making LiquidFuel is extremely difficult. A way out would be to introduce a rarer regolith variant, with much higher carbon concentration.
• To alleviate the issue, we could also decide to change the Carbon vs Hydrogen ratio of the abstracted LiquidFuel resource. Currently, it is 1:4 (CH4). Kerosene is approximately 1:2. Using a middle ground like 1:3 in conversion chains would alleviate the need for carbon, at the expanse of the (arguably more easily obtainable) hydrogen.
• I'm not sure what the exact situation is regarding obtaining hydrogen and nitrogen.

Currently, we are using the default CRP resource distributions (excepted for Nitrogen and Ammonia). CRP define a bunch of extra crustal resources : Regolith, Hydrates, Dirt, Minerals, Rock, most of them being USI or KSPIE stuff with specific purposes, abstraction levels and planetary distributions.

Worth noting is Rational Resource, which is an attempt at creating something conceptually similar as Kerbalism in terms of ISRU chains. The problem with RR is that it is fundamentally incompatible with our base abstractions. It consider LF to be LiquidHydrogen, Oxidizer to be LiquidOxygen and has a competely separate LiquidMethane chain. In doing that, it more or less entirely fail at providing compatibility with the stock and usual modding ecosystem abstractions. This being said, at it's fundamental layer, RR does something interesting with the resource distribution, reducing Ore distribution and considering it has a "high carbon content" crustal resource.

So, in essence, RR does what we need : introduce a high carbon content type of rock (ore) and provide distribution configs for it. It also does the same for Hydrates ("water-rich" crustal resource). So, the plan could be :

• Rely on the RR resource distributions (hard dependency)
• Define Ore as a carbon-rich crustal resource
• Define Hydrates as a water-rich crustal resource
• Use Regolith as the general, always present crustal resource for other purposes (do we have any ?)