MD3288: Instability latency with controlled noise

During this MD, performed on June 14th and July 26th, 2018, the latency of single bunch instabilities was tested with an external controllable noise produced by the ADT kicker. The noise acted on a subset of the bunches, whereupon one waited for those bunches to go unstable. The time interval from the moment the noise was turned on to the bunches went unstable was considered the latency of the instability. The latency has been found to be higher for either of the following: weaker noise, higher octupole current, intermediate positive chromaticity, lower transverse damper gain, lower intensity or higher emittance. Based on the bunch and machine parameters, the predicted stability threshold octupole current, in the absence of noise, was calculated with DELPHI. The growth from the predicted threshold to the octupole current when the bunches actually went unstable, per minute of active noise, has also been calculated for each bunch. This rate of change of the stability threshold octupole current has been found to be lower for either of the following: weaker noise, higher octupole current or intermediate positive chromaticity. The dependence on the transverse damper gain was inconclusive. The emittance growth rate has been analysed, and indicates that the absolute noise amplitude might have been underestimated by a factor 2.4. The ratio between the noise amplitudes applied on each subset of bunches has been controlled well.