A bench engine test has been developed which reproduces the type of turbocharger failure observed in a wide range of gasoline engined vehicles operating on commercially available API SF oils. These failures have also been reproduced in a 15000 km chassis dynamometer test developed by Audi. Initial bench engine tests showed that high temperatures alone do not cause deposit related failures. Following tests on a range of turbocharged cars a two part bench engine test was developed which discriminated between oils of different formulation. This test consisted of an initial 150 hour conditioning period run under steady state conditions followed by a cycle of wide open throttle and hot soak repeated 200 times. The combination of parameters found to be crucial in reproducing the field failures were engine oil temperature, turbocharger housing temperature, blowby, NOx and oil consumption.