In Oct 2007, John Tasker wrote to ask if steam jacketed cylinder had been considered for the 5AT. He suggests that when a loco is stationary a steam jacket would help reduce thermal losses when starting from cold. Dave Wardale offered the following response:
"Cylinder steam jacket. This has been well explained by Porta. It was rejected on the 5AT due to the difficulty of actually making a steam jacket that is effective enough to warrant the complexity - and the reduction in thickness of insulation it would result in where the latter is restricted (e.g. by the moving structure gauge). Remember it is insulation that stops heat loss, not a steam jacket - the latter just transfers any loss from cylinder steam to steam jacket steam, but it is a loss that has to be made good by the boiler in either case. One possibility is to use the valve and cylinder liner cooling passages (if fitted) as a steam jacket whenever the throttle is shut, but I doubt if it's worth the effort. It was considered for the QJ and rejected - see "The Red Devil" page 441. Another is to have a cylinder warming valve under the driver's control - or simply crack open the throttle - admitting a small amount of steam to warm up the cylinders before starting. This was used in China at very low ambient temperatures, and helps to some extent to reduce the high steam demand when starting from cold. Cylinder insulation has been covered in FDC 6  – ."