In big batches you drip in the Ketone (P2P) as fast as possible without the reaction jumping out of the flask/blowing through the condenser. A huge excess of methylamine* must be used as P2P forms the Imine not as willingly like MDP2P for example. Stirring is essential for decent yields in bigger batches. Use Al-granules about 2mm size 99,99xx% purity not foil. The amalgam can be prepared in situ, in the reaction vessel, use a generous amount of HgCl2 and warm it up to 30 °C to get this going cover the Al with alcohol + 20% water. Amalgamation works better with water added.
*4 molar or more. That's about the same weight Methylamine as P2P, more MeNH2 is better, you can recover it easily when the reaction is finished and re-use it. A efficient condenser is needed not to blow out too much methylamine with the refluxing reaction. Alternatively the top of the condenser can be connected with a tube to a vessel with water or better alcohol where the MeNH2 is collected, this avoids the horrible Amine stink too.
Ethanol (denaturated is fine) gave best results. When all P2P was added the reaction is stirred and heated to reflux until all Al is consumed. It is simple: The longer you run it the better is the yield. After finishing adding the Ketone 4 hours are absolute minimum, 5 hours are a good compromise, 12 hours improve the yields compared to 8 hours only by about 5%, longer is nonsense.
At 30 °C the reaction would have to run for days, 60 hours minimum but maybe it takes a week.
Al granules are much better and the reaction can be refluxed no problem.
Adding al P2P immediately is calling for disaster and simply retarded, why would anybody want to do this? You run it in as fast as possible, it offers perfect control, you see how far the reflux goes up the condenser and when it comes to 1/3rd you stop and wait until it calms down. Good mechanical stirring is a must have or the reaction is uncontrollable (pockets of Ketone form and react all of a sudden).
Yields are never as good as with MDP2P but with a lot of Methylamine and running it 8 hours + over 80% yield can be achieved.
Workup is adding Hydroxide and lots of water the distilling alcohol/methylamine and then the rest alcohol out of the vessel, add more water and steamdistill slowly - meth steams over very easily no need to push it, quality is much better this way. The oil from the steamdistillation can be turned into the salt directly if one did not push it, but of course further cleanup steps are always improving purity, well its up to you if you want to take pride in your product or actually do not give a fuck as its for money and for the money alone.