Nitrogen and Drying
Minimum purge velocity and the onset of stratification
A purge that moves enough total volume can still fail, because what matters is whether the interface stays sharp.
Displacement, not dilution
A good purge displaces one gas with another, maintaining a reasonably sharp interface that travels the length of the line.
A bad purge mixes them. Once mixing dominates, the purge becomes a dilution exercise, and dilution to a low concentration takes many times the line volume rather than a little over one.
Why velocity is the controlling variable
Gases of different density will separate under gravity if the flow is slow enough for buoyancy to compete with the bulk motion. Nitrogen and natural gas, or nitrogen and air, have enough density difference to stratify.
Once stratified, the lighter gas runs along the crown and the heavier along the invert, the interface smears over a long distance, and the purge stops behaving as a displacement.
Where it bites hardest
- Large-diameter lines, where the same volumetric rate gives a much lower velocity
- Rolling terrain, where gravity has a helping hand at every slope
- Low purge rates chosen to save gas
- Long runs, where a smeared interface has distance to get worse
The instinct to purge slowly to conserve nitrogen is exactly backwards: too slow and you spend far more gas reaching the endpoint.
Practical approach
Calculate the required rate from the target velocity and the internal area, rather than from the volume you intend to use:
A = π (D − 2t)² / 4 rate = velocity × A
Then confirm at the outlet. Sampling for composition at the discharge is what proves the purge, and a purge declared on calculated volume alone is an assumption.
Using a pig instead
Where the geometry allows it, a sealing pig removes the question entirely by providing a physical interface. Where it does not, velocity is the only tool you have, and it needs respecting.
From Proper Pigging, the hydrostatic and nitrogen test field book, and the HydroTech certification courses built on it.