Vacuum system model study for a 500 MeV H⁻ cyclotron Harwood, Vivienne J.; Yandon, J. C.
Using a 1 : 20 scale model of the TRIUMF vacuum chamber, simulation of chamber pumpdown performance with two different pumping system designs is described. Large trapped oil diffusion pumps and a 20°K cryopumping system, having auxilary means for pumping hydrogen, are compared in relation to the design criterion of an equivalent nitrogen stripping cross-section pressure of 7 x 10⁻⁸ Torr within 15 hours of pumping. It is shown that the design criterion can be satisfied by the cryopumping system only, when a polyurethane main chamber seal is used; for this condition resonator bakeout is not required. This result will be valid for the full-scale machine, provided the outgassing rate of the resonator surfaces does not exceed that from the model resonators by more than a factor of 3. The component outgassing rates for stainless steel with different surface treatments are discussed. Data obtained prior to and following bakeout under vacuum to 80°C and 140°C, and air baking to 150°C, are considered. The gas loads from a variety of radiation-resistant elastomers are studied; different processing treatments for polyurethane receive particular attention. A comparison is made of the predicted pumpdown performance using measured outgassing rate data and the simulated performance in the model chamber. The effect of different chamber exposure conditions on pumpdown time is described; nitrogen venting has shown to be worth while only if the exposure time to moist air is limited. A comparative study of the conductance and rate of rise methods of measuring outgassing rates and an established correlation between the rate of depletion of gas from a material and the magnitude of the pumping speed above the surface both confirm the general predictions of the surface pumping theory of Hobson and Earnshaw.
Item Citations and Data
Attribution-NonCommercial-NoDerivs 2.5 Canada