This should end the confusion:
Internal and external diameter of fused silica capillaries do not
necessarily coincide. The 'thicker' capillary may actually have the 50u
(in some cases 75u) internal diameter required for the reference gas
introduction. The flow through this capillary should be 1/16th of the
major flow only in order to maintain sensitivity. Please remember the
Poisseulle law that relates the viscous  flow through long capillaries
to the 4th power of the radius.
Regards    Willi

Willi Brand
Isotope facility of the
Max Planck Institute for Biogeochemistry, Jena Germany
<[log in to unmask]>