16-March-2001 Beam loss in the SNS linac E.Tanke, J.Galambos, J.Wei, R.Shafer, J.Stovall, T.Wangler, J.Staples, N.Catalan-Lasheras INTRODUCTION This note serves to define beam losses as sources for activation calculations in the SNS linac under normal operating conditions. Estimates for a maximum credible accident will be made at a later stage. Primary beam loss mechanisms considered here include gas stripping, and scraping due to halo and beam excursions caused by errors in linac sub-systems (e.g. steering, RF etc.). Magnetic stripping, nuclear scattering and Coulomb scattering effects have been calculated [1] to be unimportant contributors to loss for the SNS linac. BEAM LOSS DUE TO RESIDUAL GAS IN THE VACUUM In order to calculate this loss, one can assume an average beam current of 2 mA and a vacuum in the MEBT of 5x10 -7 Torr, in the DTL of 9x10 -8 Torr, in the CCL of 5x10 -8 Torr, in the SCL warm sections of 1x10 -9 Torr and 5x10 -8 Torr in the following 9 periods after the last cryomodule. The constituency of the residual gas in the DTL and CCL has been assumed to be the following RGA measured one from LEDA : H 2 18% CH 4 3% H 2 O 20% N 2 49% CO 2 7% CO 3% The residual gas in the SCL warm sections has been taken to be 100% H 2 .With the method outlined in [2], with cross sections from [3],[4], the loss rates from (single) H - stripping are found to be as shown in Table 1. The majority of the stripped H - goes to the neutral state. For the start of the DTL tank1, such a neutral would typically be lost 1.25 cm/12 mrad or ~ 1 m downstream from the point of formation (see table 1). For a residual gas pressure of 10 -7 Torr, large charge-changing cross sections for double stripping of H - to p + have been calculated for the first 30 to 50 cm of the RFQ [5],[6]. About 50 nA of protons are thus created in the RFQ and accelerated to 86 Mev (i.e. the output of the DTL) which would be finally lost in the first module of the CCL, where the RF frequency changes from 402.5 to 805 MHz. BEAM LOSS DUE TO HALO To estimate the likely locations of beam halo loss, we first look at the beam envelope relative to the aperture. Fig.2 and 3 show results from beam envelope calculations for the DTL and CCL done with TRACE3D using a matched 0.2 π mm.mrad matched beam. From it one can conclude that the DTL tank 1 and the high-energy end (>145 MeV) of the CCL have an aperture to beam size of 1 for a 6 σ beam. This ratio is about 2 for the SCL (April 2000 design) if one projects the half aperture of the warm sections all along the SCL.