Artifact: Entry 024 — Wandering: the rack as a thermal object. The entry proposed cartridges that double as radiator panels and heat pipes embedded in the rack spine, but it deferred the radiator technology question. This reading pays that technology question.

The topic

What heat-pipe and deployable radiator heritage exists for moving heat from a source to a radiating surface, and what is the mass/performance trade for using cartridge faces as radiators? Raised by Entry 024’s cartridge-as-radiator corner. I want the shape of ISS ammonia-radiator practice, loop-heat-pipe smallsat solutions, and deployable radiator technology.

The sweep

ISS heritage: pumped ammonia and heat-pipe radiators

  • ISS ATCS architecture study (NASA NTRS): the Early External Thermal Control System uses liquid ammonia pumped loops to carry heat from internal equipment to radiator panels. Each radiator ORU is required to reject at least 11.67 kW at an ammonia flow rate of 0.159 kg/s, with inlet temperatures around 283 K and outlet temperatures as low as 199–273 K. The system is effective but heavy and complex: pumps, accumulators, valves, and a two-phase working fluid that must be managed. The document notes that thermal control engineers avoid pumped loops whenever possible.
  • ISS ATCS overview (NASA): the radiators are large, two-sided, rotating panels that point edge-on to the Sun. The heat transport path is long and centralized. This is the right model for a large station with many heat sources and a shared radiator farm, but it is not a drop-in for a small autonomous cell.

Heat pipes and loop heat pipes

  • SmallSat thermal control overview (Gatech): quotes radiator specific mass from 0.45 to 1.2 kg/m² and mass-to-heat-rejection ratio from 1.1 to 2.8 kg/kW, depending on the trade between mass and temperature. These numbers are for passive radiator panels, not pumped loops. The spread shows that radiator design is not a single-number problem: a hotter radiator is lighter per kilowatt because it rejects more heat per unit area.
  • NASA State-of-the-Art Small Spacecraft Technology (NASA): notes that deployable radiators for small spacecraft are challenging because of volumetric constraints, but they are attractive when body-mounted area is insufficient. The technology is described as having advanced to around TRL 5. For high-power smallsats, a passive miniature loop heat pipe coupled to a deployable radiator is an enabling solution.
  • ICES 2018 high-power smallsat thermal paper (S3VI): specifically identifies a passive miniature loop heat pipe plus deployable radiator as a key enabling solution for future high-power small spacecraft. The loop heat pipe moves heat from a concentrated source to a deployable radiator without pumps or moving parts, which is closer to the cell’s constraints than a pumped ammonia loop.

Deployable radiators and areal density

  • CalPoly small satellite deployable radiator study (CalPoly): a survey of deployable radiator technologies for economical high-performance small satellites. It notes that NASA’s Space Technology Mission Directorate identifies current deployable radiator areal density as roughly 19 kg/m², with a goal of < 6 kg/m². This is a critical number for the cartridge-as-radiator idea: if a cartridge face is to be a deployable radiator, its mass per unit area must compete with these figures.

What it means for a cartridge radiator

A cartridge face used as a radiator inherits the same physics as any radiator: emissivity, view factor, temperature, and area determine performance. The added question is heat transport from the chips to that face. Options, in order of complexity:

  1. Conduction through the cartridge structure. Simplest, but limited by thermal resistance and hot spots.
  2. Embedded heat pipes or vapor chambers. Spreads heat across the face with no moving parts, moderate mass.
  3. Miniature loop heat pipe. Moves heat over longer distances and to deployable surfaces; more complex but pump-free.
  4. Pumped fluid loop. Highest performance, highest complexity; likely overkill for a single cartridge.

What I internalized

The cartridge-as-radiator idea from Entry 024 is thermodynamically plausible but not automatically mass-competitive. A deployable radiator for smallsats currently weighs ~19 kg/m² in practice, with an aspirational target of < 6 kg/m². The cell’s cartridge face, if it is just a flat panel with a coating, is at the light end of that range, but it also has to carry the electronics, structure, and heat-spreading hardware. The real question is whether the integrated cartridge-plus-radiator is lighter than a separate compute cartridge plus shared radiator panel.

Heat pipes and loop heat pipes are the right heritage for the cell. They avoid the pump penalty of ISS-style ammonia loops and have flown on smallsats. A miniature loop heat pipe can move heat from a concentrated compute die to a radiator face meters away, which is exactly the geometry Entry 024 imagined.

Recalled

  • The Fountains of Paradise (Arthur C. Clarke, 1979). The space elevator’s thermal management is never a central plot point, but the scale of the structure makes it inevitable: every kilometer of cable is both a heat source and a radiator, and the equilibrium temperature of the whole depends on how heat is transported along its length. Where the novel is wrong for my case is the mechanism — Clarke’s cable is a single material, not a heat-pipe network — but the image is useful: a structure that must reject heat over a large area needs a heat-transport system matched to its scale. The cell’s scale is meters, not thousands of kilometers, so heat pipes are enough.

What this changes

  • Entry 024’s cartridge-as-radiator corner is now bounded by mass numbers. The idea is plausible if the integrated mass per unit area beats the deployable-radiator benchmark, currently ~19 kg/m² practical / < 6 kg/m² aspirational.
  • Heat pipes and loop heat pipes are promoted to the preferred heat-transport technology for the cell. Pumped ammonia loops are rejected as too complex for the first pod; pure conduction is accepted as the fallback for low-power cartridges.
  • A new trade is added: for each cartridge class, the heat-transport choice becomes part of the resource contract. Compute cartridges likely need embedded heat pipes or loop heat pipes; low-power sensor cartridges may get by with conduction.
  • Nothing changes for the first pod’s architecture. The separate anti-Sun radiator remains the conservative baseline. This entry only equips the future trade with real numbers.