DISTRIBUTED GAS NETWORKS

            Many of the important systems onboard a starship require the handling of gas and cryogenic fluids, for this reason the distribution of these working fluids and fuels is essential, so wide are applications for chilled hydrogen liquid that rather than having a central store with conduits joining all systems to it, a distributed and wide spreading network is used instead.

            The principle uses for liquid hydrogen are as fuel sources, as in fusion reactors and matter antimatter reactions, and also as a cooling fluid for a wide range of equipment. Hydrogen also makes up the main constituent of most plasmas, and a source of this gas is needed to be converted into this working plasma. In short the supply of liquid hydrogen is essential for the operation of a starship, and so important that central supply would be too risky as its failure would paralyze ship operations.

            The alternative was to develop a distributed network with lots of small reservoirs scattered across the ship, these would interconnected so supplies could be routed where there was increased demand. The technology that handles liquid gases has become so well practiced that a degree of miniaturization has become possible on nearly every piece of equipment, this means that miniature gas handling apparatus can be stored in other wise dead spaces. This means that a wide network of miniaturized components can take up next to none effective volume as it can widely be installed in spaces that were unsuitable for other uses, such as spaces beneath flooring, or dead spaces in the walls, large scale equipment would have to impinge on valuable volumes, so there is great benefit in space saved.

            All the processes that are required in gas handling are serviced by large numbers of small units scattered across the network, this way no particular area has deal with a large portion of the networks handling, or if a unit fails then traffic does not have to be diverted far to another unit.

The cost of this is with many more units operating there is an increased degree of maintenance required, however failure of any particular unit does not matter in the wider context. Most components are universal units, so a defunct unit can be popped out, and an identical new one slotted into place, this kind of modular systems helps with maintenance, as if a unit appears to be operating peculiarly it can just be removed and replaced, and maintenance done on the removed unit, without risk of the disruption of service.

For major applications such as the running of a fusion reactor, which not only needs reactant but also cooling fluids, and plasma feedstocks, the network tries to act semi-autonomously for these units with partially dedicated units for these devices, so extra loads on the wider network do not increase load on these dedicated apparatus. This also means that work from these devices does not need to be dealt with on the wider network, however in the event of dedicated device failure, or local network failure, surrounding devices can adapt to compensate for their loss, which would not be possible in independently running systems.

The cost of a system with so much redundancy, is that it also becomes hideously complex, its pipes and conduits tracing throughout the ship, all of which needs to coordinated and monitored by computer systems, a substantial part of the lower networks processing time is spent optimizing gas flow in these pipes. However such a system that near enough guarantees service even in partial failure is well worth for such a valuable commodity essential for the operation of so many systems.