Experimental determination of transport coefficients, in particular internal heat transfer coefficients, in heterogeneous and hierarchical heat transfer devices such as compact regenerative heat exchangers has posed a persistent challenge for designers. The goal of this study is to (1) present a new general treatment of the experimental determination of such design data, to (2) provide simple correlations for high porosity random fiber matrices for broad design applications, and to (3) illustrate how such measurements close the formidable integro-differential volume averaging theory (VAT) equations governing transport phenomena in porous media. The combined experimental and computational method employed here for determining the internal heat transfer coefficient in the porous structure is based on the VAT model and combines with simple pressure drop measurements to yield the relevant design data for eight different high porosity random fiber samples. The design data are correlated based on a porous media length scale derived from the VAT model governing equations and the transport coefficient correlations obtained are valid for gas flows over a Reynolds number range between 5 and 70. Finally, the correlations are related to explicit, rigorously derived, lower-scale expressions arising from the VAT model. With the illustration of a new experimental tool, and the production of new simple design correlations for high porosity random fiber matrices for regenerative heat transfer applications, within the context of the hierarchical VAT model, future VAT-based simulation studies of such devices may be pursued. Moreover, the nonlocal modeling provided by VAT paves the way to meaningful optimization studies due to its singular ability to provide rigorous modeling and fast numerical solutions for transport phenomena in regenerative compact heat exchangers.