The energy transition enters the pipes: the distributed gas is no longer one, and the network calculation must handle it by composition — not by convenience coefficients.
Everything a gas «does» in a pipe — how much pressure it loses, how much it weighs on elevation differences, how it compresses — depends on three properties: relative density, dynamic viscosity and compressibility. The robust way to obtain them, valid for any gas, is to compute them from the mixture's molar composition: molar masses for the density, mixing rules for the viscosity, pseudo-critical properties for the compressibility. This way natural gas, biomethane, LPG and hydrogen blends are the same problem with different numbers — and no «special correlations» to justify.
Grid-injected biomethane is high-purity methane: the injection quality specification imposes a composition very close to natural gas, precisely so that the network and the appliances do not notice the difference. Hydraulically the effect is therefore modest — but the correct calculation remains the compositional one: declare the real mixture and let the properties follow.
Hydrogen is another story: relative density ≈ 0.07 (against ≈ 0.55 for pure methane), lower viscosity, volumetric heating value about one third. In a blend the properties shift proportionally, and the two main effects are:
There is also a second-order effect that quick calculations ignore: in hydrogen-rich blends the mixture's viscosity does not follow a naive average, and a proper mixing rule (Herning–Zipperer) is needed for the Reynolds number — and hence the friction — to stay right.
The real biomethane-injection scenario is not «the whole network switches gas»: it is a network with several sources injecting different gases — the delivery from the transmission grid on one side, the local production plant on the other. In between, zones served by one, the other, or both. A correct calculation apportions the gases across the network and assigns each pipe the properties of the gas that actually flows through it, instead of smearing an average mixture everywhere.
Gasnetics works this way: free molar composition for the design mixture, viscosity correction for H₂-rich blends, the altitude correction with the right sign for any density, and — with several sources — a mixture of its own for each, with the gas apportionment solved together with the hydraulic balance.
Little: grid-injected biomethane is high-purity methane by quality specification, with properties very close to natural gas. The correct treatment is still by molar composition: declare the real mixture and let density, viscosity and compressibility follow from it.
Hydrogen is much lighter than methane (relative density about 0.07 against 0.55): it lowers the blend's density and volumetric heating value. At equal volumetric flow the drops decrease, but at equal delivered energy more volume is needed — about three times for pure hydrogen — so the hydraulic balance must be recomputed, not extrapolated.
That is the typical biomethane-injection scenario: one delivery point with grid gas and a local plant with its own gas. A correct calculation apportions the gases across the network and assigns each pipe the properties of the gas it actually carries, instead of assuming one mixture everywhere.