In 1988, a prototypical vacuum system was added to the Tritium Systems Test Assembly (TSTA) at Los Alamos National Laboratory. Since then various pumping scenarios, which might be expected in a fusion reactor, have been performed without any serious shortcomings being apparent in the use of compound cryopumps as reactor high vacuum pumps. Last year, the question of whether a compound pump was necessary was addressed in a pair of runs in which deuterium helium mixtures were pumped on a single 4K activated charcoal panel. In these tests, the condensing stage of the pump was maintained at 77K and did not contribute to pumping either deuterium or helium. Results were very encouraging: in both tests the charcoal readily pumped helium until a max loading of 0.4 T 1 cm−2 of helium on charcoal was attained. Helium speed was not affected by deuterium which may have been pumped by either a condensing or sorbing mechanism or by a combination of both. In addition, the helium loading at saturation was 0.4 T 1 cm−2 even though the D2/He ratio was doubled between runs. Conjecture about why the charcoal helium capacity was constant led to the pump operation described in this paper. It was felt that measurement of helium capacity after careful deuterium preloads might help to explain the mechanism involved in co-pumping of a condensible and a noncondensible on a single 4K cryosorber surface. This paper presents the results of series of helium capacity runs preceded by a range of deuterium preloads and attempts to explain the mechanism involved.